# for emacs: -*- mode: sh; -*-

# This file describes browser build for the mm10
# Mus musculus (mouse)

#	DATE:   07-Dec-2011
#	ORGANISM:       Mus musculus
#	TAXID:  10090
#	ASSEMBLY LONG NAME:     Genome Reference Consortium Mouse Build 38
#	ASSEMBLY SHORT NAME:    GRCm38
#	ASSEMBLY SUBMITTER:     Genome Reference Consortium
#	ASSEMBLY TYPE:  Haploid + alternate loci
#	NUMBER OF ASSEMBLY-UNITS:       16
#	ASSEMBLY ACCESSION:     GCA_000001635.2

#	rsync://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Mus_musculus/GRCm38/

#	Genome ID:
#	http://www.ncbi.nlm.nih.gov/genome/52

#	Taxonomy:
#	http://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?id=10090
#	http://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?id=39442

#	GRC information
#	http://www.ncbi.nlm.nih.gov/projects/genome/assembly/grc/mouse/

#	Mitochondrial sequence:
#	http://www.ncbi.nlm.nih.gov/bioproject/13767
#	C57BL/6J sequence:
#	http://www.ncbi.nlm.nih.gov/bioproject/51977
#	Finishing project:
#	http://www.ncbi.nlm.nih.gov/bioproject/20689

#	Assembly ID: 327618
#	http://www.ncbi.nlm.nih.gov/genome/assembly/327618/

#	Celera Assembly
# http://www.ncbi.nlm.nih.gov/Traces/wgs/?val=AAHY00

#############################################################################
# fetch sequence from genbank (DONE - 2012-01-30 - Hiram)
    mkdir -p /hive/data/genomes/mm10/genbank
    cd /hive/data/genomes/mm10/genbank

    rsync -a -P \
rsync://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Mus_musculus/GRCm38/ ./

    # measure sequence to be used here
    faSize Primary_Assembly/assembled_chromosomes/FASTA/*.fa.gz \
	Primary_Assembly/unplaced_scaffolds/FASTA/*.fa.gz \
	Primary_Assembly/unlocalized_scaffolds/FASTA/*.fa.gz \
	non-nuclear/assembled_chromosomes/FASTA/chrMT.fa.gz
    # 2730871774 bases (78088274 N's 2652783500 real 2652783500 upper 0 lower)
    # in 66 sequences in 29 files
    #	Total size: mean 41376845.1 sd 63617337.3 min 1976
    #	(gi|371559559|gb|JH584295.1|) max 195471971
    #	(gi|371561115|gb|CM000994.2|) median 184189

#############################################################################
# fixup names for UCSC standards (DONE - 2012-02-06 - Hiram)
    mkdir /hive/data/genomes/mm10/ucsc
    cd /hive/data/genomes/mm10/ucsc

    ########################  Assembled Chromosomes
    cat << '_EOF_' > toUcsc.pl
#!/bin/env perl

use strict;
use warnings;

my %accToChr;

open (FH, "<../genbank/Primary_Assembly/assembled_chromosomes/chr2acc") or
        die "can not read Primary_Assembly/assembled_chromosomes/chr2acc";
while (my $line = <FH>) {
    next if ($line =~ m/^#/);
    chomp $line;
    my ($chrN, $acc) = split('\s+', $line);
    $accToChr{$acc} = $chrN;
}
close (FH);

foreach my $acc (keys %accToChr) {
    my $chrN =  $accToChr{$acc};
    print "$acc $accToChr{$acc}\n";
    open (FH, "zcat ../genbank/Primary_Assembly/assembled_chromosomes/AGP/chr${chrN}.agp.gz|") or die "can not read chr${chrN}.agp.gz";
    open (UC, ">chr${chrN}.agp") or die "can not write to chr${chrN}.agp";
    while (my $line = <FH>) {
        if ($line =~ m/^#/) {
            print UC $line;
        } else {
            $line =~ s/^$acc/chr${chrN}/;
            print UC $line;
        }
    }
    close (FH);
    close (UC);
    open (FH, "zcat ../genbank/Primary_Assembly/assembled_chromosomes/FASTA/chr${chrN}.fa.gz|") or die "can not read chr${chrN}.fa.gz";
    open (UC, ">chr${chrN}.fa") or die "can not write to chr${chrN}.fa";
    while (my $line = <FH>) {
        if ($line =~ m/^>/) {
            printf UC ">chr${chrN}\n";
        } else {
            print UC $line;
        }
    }
    close (FH);
    close (UC);
}
'_EOF_'
    # << happy emacs
    chmod +x toUcsc.pl
    time ./toUcsc.pl
    #	real    0m53.256s
    faSize chr*.fa
    #	2725521370 bases (77999939 N's 2647521431 real 2647521431 upper 0
    #	lower) in 21 sequences in 21 files
    #	Total size: mean 129786731.9 sd 33408399.1 min 61431566 (chr19)
    #	max 195471971 (chr1) median 124902244

    ########################  Unplaced scaffolds
    cat << '_EOF_' > unplaced.pl
#!/bin/env perl

use strict;
use warnings;

my $agpFile =  "../genbank/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz";
my $fastaFile =  "../genbank/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fa.gz";
open (FH, "zcat $agpFile|") or die "can not read $agpFile";
open (UC, ">unplaced.agp") or die "can not write to unplaced.agp";
while (my $line = <FH>) {
    if ($line =~ m/^#/) {
        print UC $line;
    } else {
        $line =~ s/\.1//;
        printf UC "chrUn_%s", $line;
    }
}
close (FH);
close (UC);

open (FH, "zcat $fastaFile|") or die "can not read $fastaFile";
open (UC, ">unplaced.fa") or die "can not write to unplaced.fa";
while (my $line = <FH>) {
    if ($line =~ m/^>/) {
        chomp $line;
        $line =~ s/.*gb\|//;
        $line =~ s/\.1\|.*//;
        printf UC ">chrUn_$line\n";
    } else {
        print UC $line;
    }
}
close (FH);
close (UC);
'_EOF_'
    # << happy emacs
    chmod +x unplaced.pl
    time ./unplaced.pl
    #	real    0m0.119s
    # make sure none of the names got to be over 31 characers long:
    grep -v "^#" unplaced.agp | cut -f1 | sort | uniq -c | sort -rn
    # not much in that sequence:
    faSize unplaced.fa
    #	803895 bases (62411 N's 741484 real 741484 upper 0 lower)
    #	in 22 sequences in 1 files
    #	Total size: mean 36540.7 sd 21518.0 min 20208 (chrUn_GL456368)
    #	max 114452 (chrUn_JH584304) median 28772

    ########## chrM
    zcat ../genbank/non-nuclear/assembled_chromosomes/FASTA/chrMT.fa.gz \
	| sed -e "s/^>.*/>chrM/" > chrM.fa
    zcat ../genbank/non-nuclear/assembled_chromosomes/AGP/chrMT.comp.agp.gz \
	| sed -e "s/^AY172335.1/chrM/" > chrM.agp

    ########################  Unlocalized scaffolds
    cat << '_EOF_' > unlocalized.pl
#!/bin/env perl

use strict;
use warnings;

my %accToChr;
my %chrNames;

open (FH, "<../genbank/Primary_Assembly/unlocalized_scaffolds/unlocalized.chr2scaf") or
        die "can not read Primary_Assembly/unlocalized_scaffolds/unlocalized.chr2scaf";
while (my $line = <FH>) {
    next if ($line =~ m/^#/);
    chomp $line;
    my ($chrN, $acc) = split('\s+', $line);
    $accToChr{$acc} = $chrN;
    $chrNames{$chrN} += 1;
}
close (FH);

foreach my $chrN (keys %chrNames) {
    my $agpFile =  "../genbank/Primary_Assembly/unlocalized_scaffolds/AGP/chr$chrN.unlocalized.scaf.agp.gz";
    my $fastaFile =  "../genbank/Primary_Assembly/unlocalized_scaffolds/FASTA/chr$chrN.unlocalized.scaf.fa.gz";
    open (FH, "zcat $agpFile|") or die "can not read $agpFile";
    open (UC, ">chr${chrN}_random.agp") or die "can not write to chr${chrN}_random.agp";
    while (my $line = <FH>) {
        if ($line =~ m/^#/) {
            print UC $line;
        } else {
            chomp $line;
            my (@a) = split('\t', $line);
            my $acc = $a[0];
            my $accNo1 = $acc;
            $accNo1 =~ s/.1$//;
            die "ERROR: acc not .1: $acc" if ($accNo1 =~ m/\./);
            die "ERROR: chrN $chrN not correct for $acc"
                if ($accToChr{$acc} ne $chrN);
            my $ucscName = "chr${chrN}_${accNo1}_random";
            printf UC "%s", $ucscName;
            for (my $i = 1; $i < scalar(@a); ++$i) {
                printf UC "\t%s", $a[$i];
            }
            printf UC "\n";
        }
    }
    close (FH);
    close (UC);
    printf "chr%s\n", $chrN;
    open (FH, "zcat $fastaFile|") or die "can not read $fastaFile";
    open (UC, ">chr${chrN}_random.fa") or die "can not write to chr${chrN}_random.fa";
    while (my $line = <FH>) {
        if ($line =~ m/^>/) {
            chomp $line;
            my $acc = $line;
            $acc =~ s/.*gb\|//;
            $acc =~ s/\|.*//;
            my $accNo1 = $acc;
            $accNo1 =~ s/.1$//;
            die "ERROR: acc not .1: $acc" if ($accNo1 =~ m/\./);
            die "ERROR: chrN $chrN not correct for $acc"
                if ($accToChr{$acc} ne $chrN);
            my $ucscName = "chr${chrN}_${accNo1}_random";
            printf UC ">$ucscName\n";
        } else {
            print UC $line;
        }
    }
    close (FH);
    close (UC);
}
'_EOF_'
    # << happy emacs
    chmod +x unlocalized.pl
    time ./unlocalized.pl
    #	real    0m0.430s
    faSize chr*_random.fa
    #	4530210 bases (25924 N's 4504286 real 4504286 upper 0 lower)
    #	in 22 sequences in 6 files
    #	Total size: mean 205918.6 sd 184688.0 min 1976 (chr4_JH584295_random)
    #	max 953012 (chr5_JH584299_random) median 191905
    # verify none of the names are longer than 31 characters:
    grep -h -v "^#" chr*_random.agp | cut -f1 | sort | uniq -c | sort -nr


    # compress all these fasta and agp files:
    gzip *.fa *.agp
    #	verify all the sequence is still here after all this rigamarole:
    time faSize *.fa.gz
    #	2730871774 bases (78088274 N's 2652783500 real 2652783500 upper 0
    #	lower) in 66 sequences in 29 files
    #	Total size: mean 41376845.1 sd 63617337.3 min 1976
    #	(chr4_JH584295_random) max 195471971 (chr1) median 184189

#############################################################################
#   Initial browser build (DONE - 2012-01-06 - Hiram)
    cd /hive/data/genomes/mm10
    cat << '_EOF_' > mm10.config.ra
# Config parameters for makeGenomeDb.pl:
db mm10
clade mammal
genomeCladePriority 40
scientificName Mus musculus
commonName Mouse
assemblyDate Dec. 2011
assemblyLabel Genome Reference Consortium Mouse Build 38 (GCA_000001635.2)
assemblyShortLabel GRCm38
orderKey 1209
mitoAcc none
fastaFiles /hive/data/genomes/mm10/ucsc/*.fa.gz
agpFiles /hive/data/genomes/mm10/ucsc/*.agp.gz
dbDbSpeciesDir mouse
taxId   10090
ncbiAssemblyId   327618
ncbiAssemblyName   GRCm38
'_EOF_'
    # << happy emacs

    time makeGenomeDb.pl -stop=agp mm10.config.ra > agp.log 2>&1
    #	real    3m4.568s
    # check the end of agp.log to verify it is OK
    time makeGenomeDb.pl -workhorse=hgwdev -fileServer=hgwdev \
	-continue=db mm10.config.ra > db.log 2>&1
    #	real    20m51.374s
    # verify the end of db.log indicates successful


#############################################################################
# running repeat masker (DONE - 2012-02-06 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/repeatMasker
    cd /hive/data/genomes/mm10/bed/repeatMasker
    time doRepeatMasker.pl -buildDir=`pwd` -noSplit \
	-bigClusterHub=swarm -dbHost=hgwdev -workhorse=hgwdev \
	-smallClusterHub=encodek mm10 > do.log 2>&1 &
    #	real    609m48.767s

    cat faSize.rmsk.txt
    #	2730871774 bases (78088274 N's 2652783500 real 1456094545 upper
    #	1196688955 lower) in 66 sequences in 1 files
    #	Total size: mean 41376845.1 sd 63617337.3 min 1976
    #	(chr4_JH584295_random) max 195471971 (chr1) median 184189
    #	%43.82 masked total, %45.11 masked real

    grep -i versi do.log
# RepeatMasker version development-$Id: RepeatMasker,v 1.26 2011/09/26 16:19:44 angie Exp $
#    April 26 2011 (open-3-3-0) version of RepeatMasker

    time featureBits -countGaps mm10 rmsk
    #	1196694219 bases of 2730871774 (43.821%) in intersection
    #	real    0m30.460s
    # why is it different than the faSize above ?
    # because rmsk masks out some N's as well as bases, the count above
    #	separates out the N's from the bases, it doesn't show lower case N's

##########################################################################
# running simple repeat (DONE - 2012-02-06 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/simpleRepeat
    cd /hive/data/genomes/mm10/bed/simpleRepeat
    time doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=swarm \
	-dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=encodek \
	mm10 > do.log 2>&1 &
    #	real    16m35.603s

    #	batch failed, one job failed:
    # ./TrfRun.csh /hive/data/genomes/mm10/TrfPart/062/062.lst.bed
    # which is the chrM sequence - it has no simple repeats
    # create an empty output file result:
    touch /hive/data/genomes/mm10/TrfPart/062/062.lst.bed
    # go to encodek and create the run.time file to signal this step is done
    cd /hive/data/genomes/mm10/bed/simpleRepeat/run.cluster
    para time > run.time
# Completed: 70 of 71 jobs
# Crashed: 1 jobs
# CPU time in finished jobs:      13103s     218.38m     3.64h    0.15d  0.000 y
# IO & Wait Time:                   163s       2.72m     0.05h    0.00d  0.000 y
# Average job time:                 190s       3.16m     0.05h    0.00d
# Longest finished job:             392s       6.53m     0.11h    0.00d
# Submission to last job:           894s      14.90m     0.25h    0.01d


    # continue procedure:
    time doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=swarm \
	-dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=encodek \
	-continue=filter mm10 > filter.log 2>&1 &
    #	real    1m20.021s

    cat fb.simpleRepeat
    #	92161833 bases of 2652783500 (3.474%) in intersection

    # when RepeatMasker is done, add this mask to the sequence:
    cd /hive/data/genomes/mm10
    twoBitMask mm10.rmsk.2bit \
	-add bed/simpleRepeat/trfMask.bed mm10.2bit
    #	you can safely ignore the warning about fields >= 13

    twoBitToFa mm10.2bit stdout | faSize stdin > faSize.mm10.2bit.txt
    cat faSize.mm10.2bit.txt
    #	2730871774 bases (78088274 N's 2652783500 real 1454267808 upper
    #	1198515692 lower) in 66 sequences in 1 files
    #	Total size: mean 41376845.1 sd 63617337.3 min 1976
    #	(chr4_JH584295_random) max 195471971 (chr1) median 184189
    #	%43.89 masked total, %45.18 masked real

    # set SymLink in gbdb to this masked sequence
    rm /gbdb/mm10/mm10.2bit
    ln -s `pwd`/mm10.2bit /gbdb/mm10/mm10.2bit

#########################################################################
# Verify all gaps are marked, add any N's not in gap as type 'other'
#	(DONE - 2012-02-06 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/gap
    cd /hive/data/genomes/mm10/bed/gap
    time nice -n +19 findMotif -motif=gattaca -verbose=4 \
	-strand=+ ../../mm10.unmasked.2bit > findMotif.txt 2>&1
    #	real    1m0.372s
    grep "^#GAP " findMotif.txt | sed -e "s/^#GAP //" > allGaps.bed
    time featureBits -countGaps mm10 -not gap -bed=notGap.bed
    #	2658879040 bases of 2730871774 (97.364%) in intersection
    #	real    0m13.067s

    time featureBits -countGaps mm10 allGaps.bed notGap.bed -bed=new.gaps.bed
    #	6095540 bases of 2730871774 (0.223%) in intersection
    #	real    0m15.177s

    #	what is the highest index in the existing gap table:
    hgsql -N -e "select ix from gap;" mm10 | sort -n | tail -1
    #	54
    cat << '_EOF_' > mkGap.pl
#!/bin/env perl

use strict;
use warnings;

my $ix=`hgsql -N -e "select ix from gap;" mm10 | sort -n | tail -1`;
chomp $ix;

open (FH,"<new.gaps.bed") or die "can not read new.gaps.bed";
while (my $line = <FH>) {
    my ($chrom, $chromStart, $chromEnd, $rest) = split('\s+', $line);
    ++$ix;
    printf "%s\t%d\t%d\t%d\tN\t%d\tother\tyes\n", $chrom, $chromStart,
        $chromEnd, $ix, $chromEnd-$chromStart;
}
close (FH);
'_EOF_'
    # << happy emacs
    chmod +x ./mkGap.pl
    ./mkGap.pl > other.bed
    wc -l other.bed
    #	384
    featureBits -countGaps mm10 other.bed
    #	6095540 bases of 2730871774 (0.223%) in intersection
    hgLoadBed -sqlTable=$HOME/kent/src/hg/lib/gap.sql \
	-noLoad mm10 otherGap other.bed
    # verify no overlap with gap table:
    time featureBits -countGaps mm10 gap other.bed
    #	0 bases of 2730871774 (0.000%) in intersection
    #	real    0m1.281s

    # verify no errors before adding to the table:
    time gapToLift -minGap=1 mm10 nonBridged.before.lift \
	-bedFile=nonBridged.before.bed > before.gapToLift.txt 2>&1 &
    #	real    0m7.205s
    # check for warnings in before.gapToLift.txt, should be empty:
    #	-rw-rw-r-- 1     1633 Jan  6 15:20 before.gapToLift.txt
    # it indicates that there are telomere's adjacent to centromere's
    #	and heterochromatin
    #	starting with this many:
    hgsql -e "select count(*) from gap;" mm10
    #	302
    hgsql mm10 -e 'load data/genomes local infile "bed.tab" into table gap;'
    #	result count:
    hgsql -e "select count(*) from gap;" mm10
    #	686
    # == 302 + 384
    # verify we aren't adding gaps where gaps already exist
    # this would output errors if that were true:
    gapToLift -minGap=1 mm10 nonBridged.lift -bedFile=nonBridged.bed
    #same set of warnings as before, telomere's centromere's and heterochromatin
    # there should be no errors or other output, checked bridged gaps:
    hgsql -N -e "select bridge from gap;" mm10 | sort | uniq -c
    #	191 no
    #	495 yes

##########################################################################
## WINDOWMASKER (DONE - 2012-02-06 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/windowMasker
    cd /hive/data/genomes/mm10/bed/windowMasker
    time nice -n +19 doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \
	-dbHost=hgwdev mm10 > do.log 2>&1 &
    #	real    167m12.012s

    # Masking statistics
    twoBitToFa mm10.wmsk.2bit stdout | faSize stdin
    #	2730871774 bases (78088274 N's 2652783500 real 1686407708 upper
    #	966375792 lower) in 66 sequences in 1 files
    #	Total size: mean 41376845.1 sd 63617337.3 min 1976
    #	(chr4_JH584295_random) max 195471971 (chr1) median 184189
    #	%35.39 masked total, %36.43 masked real


    twoBitToFa mm10.wmsk.sdust.2bit stdout | faSize stdin
    #	2730871774 bases (78088274 N's 2652783500 real 1670424648 upper
    #	982358852 lower) in 66 sequences in 1 files
    #	Total size: mean 41376845.1 sd 63617337.3 min 1976
    #	(chr4_JH584295_random) max 195471971 (chr1) median 184189
    #	%35.97 masked total, %37.03 masked real

    hgLoadBed mm10 windowmaskerSdust windowmasker.sdust.bed.gz
    #	Loaded 12655947 elements of size 3

    featureBits -countGaps mm10 windowmaskerSdust
    #	1060447084 bases of 2730871774 (38.832%) in intersection

    #	eliminate the gaps from the masking
    featureBits mm10 -not gap -bed=notGap.bed
    #	2652783500 bases of 2652783500 (100.000%) in intersection
    time nice -n +19 featureBits mm10 windowmaskerSdust notGap.bed \
        -bed=stdout | gzip -c > cleanWMask.bed.gz
    #	982358852 bases of 2652783500 (37.031%) in intersection
    #	real    1m42.449s

    #	reload track to get it clean
    hgLoadBed mm10 windowmaskerSdust cleanWMask.bed.gz
    #	Loaded 12655987  elements of size 4
    time featureBits -countGaps mm10 windowmaskerSdust
    #	982358852 bases of 2730871774 (35.972%) in intersection
    #	real    1m13.889s

    #	do *not* need to mask with this clean result since RepeatMasker
    #	does a very good job here.  Using RM masking instead.
    zcat cleanWMask.bed.gz \
	| twoBitMask ../../mm10.unmasked.2bit stdin \
	    -type=.bed mm10.cleanWMSdust.2bit
    twoBitToFa mm10.cleanWMSdust.2bit stdout | faSize stdin \
        > mm10.cleanWMSdust.faSize.txt
    cat mm10.cleanWMSdust.faSize.txt

    # how much does this window masker and repeat masker overlap:
    time featureBits -countGaps mm10 rmsk windowmaskerSdust
    #	753614881 bases of 2730871774 (27.596%) in intersection
    #	real    1m42.691s
    # RM by itself:
    time featureBits -countGaps mm10 rmsk
    #	1196694219 bases of 2730871774 (43.821%) in intersection
    #	real    0m30.460s

#############################################################################
# PREPARE LINEAGE SPECIFIC REPEAT FILES FOR BLASTZ (DONE - 2012-02-07 - Hiram)
    ssh encodek
    mkdir /hive/data/genomes/mm10/bed/linSpecRep
    cd /hive/data/genomes/mm10/bed/linSpecRep

    # split the RM output by chromosome name into separate files
    mkdir rmsk dateRepeats
    head -3 ../repeatMasker/mm10.sorted.fa.out > rmsk.header.txt
    headRest 3 ../repeatMasker/mm10.sorted.fa.out \
	| splitFileByColumn -ending=.out -col=5 -head=rmsk.header.txt stdin rmsk

    ls -1S rmsk/* > rmOut.list

    cat << '_EOF_' > mkLSR
#!/bin/csh -fe
rm -f dateRepeats/$1_homo-sapiens_rattus_canis-familiaris_bos-taurus
/scratch/data/genomes/RepeatMasker/DateRepeats \
    $1 -query mouse -comp human -comp rat -comp dog -comp cow
mv $1_homo-sapiens_rattus_canis-lupus-familiaris_bos-taurus dateRepeats
'_EOF_'
    #	<< happy emacs
    chmod +x mkLSR

    cat << '_EOF_' > template
#LOOP
./mkLSR $(path1) {check out line+ dateRepeats/$(file1)_homo-sapiens_rattus_canis-lupus-familiaris_bos-taurus}
#ENDLOOP
'_EOF_'
    #	<< happy emacs

    gensub2 rmOut.list single template jobList
    para create jobList
    para try ... check ... push ... etc...
    para time
# Completed: 66 of 66 jobs
# CPU time in finished jobs:       1743s      29.05m     0.48h    0.02d  0.000 y
# IO & Wait Time:                   190s       3.16m     0.05h    0.00d  0.000 y
# Average job time:                  29s       0.49m     0.01h    0.00d
# Longest finished job:              65s       1.08m     0.02h    0.00d
# Submission to last job:           160s       2.67m     0.04h    0.00d

    mkdir notInHuman notInRat notInDog notInCow
    for F in dateRepeats/chr*.out_homo-sapiens*
    do
	B=`basename ${F}`
	B=${B/.out*/}
	echo $B
        /cluster/bin/scripts/extractRepeats 1 ${F} > \
		notInHuman/${B}.out.spec
        /cluster/bin/scripts/extractRepeats 2 ${F} > \
		notInRat/${B}.out.spec
        /cluster/bin/scripts/extractRepeats 3 ${F} > \
		notInDog/${B}.out.spec
        /cluster/bin/scripts/extractRepeats 4 ${F} > \
		notInCow/${B}.out.spec
    done

    #	notInDog, and notInCow ended up being identical.
    #	The notInRat and notInHuman are different
    #	To check identical
    find . -name "*.out.spec" | \
	while read FN; do echo `cat ${FN} | sum -r` ${FN}; done \
	| sort -k1,1n | sort -t"/" -k3,3 > check.same
    # this produces a count of 2 for the sums for Cow and Dog, all the same
    egrep "Cow|Dog" check.same | awk '{print $1}' | sort | uniq -c | sort -rn
    # this does not produce a count of 2 for the sums for Cow and Human
    egrep "Cow|Human" check.same | awk '{print $1}' | sort | uniq -c | sort -rn
    #	Copy to data/genomes staging for cluster replication
    mkdir /hive/data/genomes/staging/data/genomes/mm10
    rsync -a -P ./notInRat/ /hive/data/genomes/staging/data/genomes/mm10/notInRat/
    rsync -a -P ./notInHuman/ /hive/data/genomes/staging/data/genomes/mm10/notInHuman/
    rsync -a -P ./notInCow/ /hive/data/genomes/staging/data/genomes/mm10/notInOthers/


    # We also need the nibs for the lastz runs with lineage specific repeats
    mkdir /hive/data/genomes/mm10/nib
    cd /hive/data/genomes/mm10
    cut -f1 chrom.sizes | while read C
do
    twoBitToFa -seq=${C} mm10.2bit stdout | faToNib -softMask stdin nib/${C}.nib
    ls -og nib/$C.nib
done
    # verify one is properly masked:
    nibFrag -masked nib/chrM.nib 0 16299 + stdout | less
    # compare to:
    twoBitToFa -seq=chrM mm10.fa stdout | less

    #	Copy to data/genomes staging for cluster replication
    rsync -a -P ./nib/ /hive/data/genomes/staging/data/genomes/mm10/nib/

#########################################################################
# MAKE 11.OOC FILE FOR BLAT/GENBANK (DONE - 2012-02-08 - Hiram)
    # Use -repMatch=650, based on size -- for human we use 1024
    # use the "real" number from the faSize measurement,
    # hg19 is 2897316137, calculate the ratio factor for 1024:
    calc \( 2652783500 / 2897316137 \) \* 1024
    #	( 2652783500 / 2897316137 ) * 1024 = 937.574699

    # round up to 1000  (mm9 used 912)

    cd /hive/data/genomes/mm10
    time blat mm10.2bit /dev/null /dev/null -tileSize=11 \
      -makeOoc=jkStuff/mm10.11.ooc -repMatch=1000
    #	Wrote 27208 overused 11-mers to jkStuff/mm10.11.ooc
    #	real    2m9.568s

    #	at repMatch=900:
    #	Wrote 31822 overused 11-mers to jkStuff/mm10.11.ooc

    # there are non-bridged gaps, make lift file for genbank
    hgsql -N -e "select bridge from gap;" mm10 | sort | uniq -c
    #	191 no
    #	495 yes
    cd /hive/data/genomes/mm10/jkStuff
    gapToLift mm10 mm10.nonBridged.lift -bedFile=mm10.nonBridged.bed
    # largest non-bridged contig:
    awk '{print $3-$2,$0}' mm10.nonBridged.bed | sort -nr | head
    116378660 chr2  59120641        175499301       chr2.02

    #	copy all of this stuff to the klusters:
    cd /hive/data/genomes/mm10
    mkdir /hive/data/genomes/staging/data/genomes/mm10
    cp -p jkStuff/mm10.11.ooc jkStuff/mm10.nonBridged.lift chrom.sizes \
	mm10.2bit /hive/data/genomes/staging/data/genomes/mm10
    # request rsync copy from cluster admin

#########################################################################
# AUTO UPDATE GENBANK (DONE - 2012-02-08 - Hiram)
    # examine the file:
    /cluster/data/genomes/genbank/data/genomes/organism.lst
    # for your species to see what counts it has for:
# organism       mrnaCnt estCnt  refSeqCnt
# Mus musculus    334577  4853663 26288
    # to decide which "native" mrna or ests you want to specify in genbank.conf
    # of course, mm10 has plenty of everything

    ssh hgwdev
    cd $HOME/kent/src/hg/makeDb/genbank
    git pull
    # edit etc/genbank.conf to add mm10 just after mm9 and commit to GIT
# mm10
mm10.serverGenome = /hive/data/genomes/mm10/mm10.2bit
mm10.clusterGenome = /scratch/data/genomes/mm10/mm10.2bit
mm10.ooc = /scratch/data/genomes/mm10/mm10.11.ooc
mm10.align.unplacedChroms = chr*
mm10.lift = /scratch/data/genomes/mm10/mm10.nonBridged.lift
mm10.refseq.mrna.native.pslCDnaFilter  = ${finished.refseq.mrna.native.pslCDnaFilter}
mm10.refseq.mrna.xeno.pslCDnaFilter    = ${finished.refseq.mrna.xeno.pslCDnaFilter}
mm10.genbank.mrna.native.pslCDnaFilter = ${finished.genbank.mrna.native.pslCDnaFilter}
mm10.genbank.mrna.xeno.pslCDnaFilter   = ${finished.genbank.mrna.xeno.pslCDnaFilter}
mm10.genbank.est.native.pslCDnaFilter  = ${finished.genbank.est.native.pslCDnaFilter}
mm10.downloadDir = mm10
mm10.refseq.mrna.xeno.load  = yes
mm10.refseq.mrna.xeno.loadDesc = yes
mm10.mgc = yes
mm10.genbank.mrna.blatTargetDb = yes
# mm10.ccds.ncbiBuild = 37.2
# mm10.upstreamGeneTbl = refGene
# mm10.upstreamMaf = multiz30way
# /hive/data/genomes/mm10/bed/multiz30way/species.list

    # end of section added to etc/genbank.conf
    git commit -m "adding mm10 definitions" genbank.conf
    git push
    make etc-update

    ssh hgwdev			# used to do this on "genbank" machine
    screen			# long running job managed in screen
    cd /cluster/data/genomes/genbank
    time nice -n +19 ./bin/gbAlignStep -initial mm10 &
    #	var/build/logs/2012.02.08-11:38:50.mm10.initalign.log
    #	real    795m52.388s

    # load data/genomesbase when finished
    ssh hgwdev
    cd /cluster/data/genomes/genbank
    time nice -n +19 ./bin/gbDbLoadStep -drop -initialLoad mm10 &
    #	logFile: var/dbload/hgwdev/logs/2012.02.09-10:05:25.dbload.log
    #	real    114m56.461s

    # enable daily alignment and update of hgwdev (DONE - 2012-02-09 - Hiram)
    cd ~/kent/src/hg/makeDb/genbank
    git pull
    # add mm10 to:
        etc/align.dbs
        etc/hgwdev.dbs
    git commit -m "Added mm10." etc/align.dbs etc/hgwdev.dbs
    git push
    make etc-update

############################################################################
# running cpgIsland business (DONE - 2012-02-09 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/cpgIsland
    cd /hive/data/genomes/mm10/bed/cpgIsland
    # use a previous binary for this program
    ln -s ../../../mm9/bed/cpgIsland/hg3rdParty/cpgIslands/cpglh.exe .

    mkdir -p hardMaskedFa
    cut -f1 ../../chrom.sizes | while read C
do
    echo ${C}
    twoBitToFa ../../mm10.2bit:$C stdout \
	| maskOutFa stdin hard hardMaskedFa/${C}.fa
done

    ssh swarm
    cd /hive/data/genomes/mm10/bed/cpgIsland
    mkdir results
    cut -f1 ../../chrom.sizes > chr.list
    cat << '_EOF_' > template
#LOOP
./runOne $(root1) {check out exists results/$(root1).cpg}
#ENDLOOP
'_EOF_'
    # << happy emacs

    #	the faCount business is to make sure there is enough sequence to
    #	work with in the fasta.  cpglh.exe does not like files with too many
    #	N's - it gets stuck.
    cat << '_EOF_' > runOne
#!/bin/csh -fe
set C = `faCount hardMaskedFa/$1.fa | egrep -v "^#seq|^total" | awk '{print  $2 - $7 }'`
if ( $C > 200 ) then
    ./cpglh.exe hardMaskedFa/$1.fa > /scratch/tmp/$1.$$
    mv /scratch/tmp/$1.$$ $2
else
    touch $2
endif
'_EOF_'
    # << happy emacs
    chmod +x runOne

    gensub2 chr.list single template jobList
    para create jobList
    para try
    para check ... etc
    para time
# Completed: 66 of 66 jobs
# CPU time in finished jobs:        191s       3.19m     0.05h    0.00d  0.000 y
# IO & Wait Time:                   189s       3.14m     0.05h    0.00d  0.000 y
# Average job time:                   6s       0.10m     0.00h    0.00d
# Longest finished job:              19s       0.32m     0.01h    0.00d
# Submission to last job:            51s       0.85m     0.01h    0.00d

    # Transform cpglh output to bed +
    catDir results | awk '{
$2 = $2 - 1;
width = $3 - $2;
printf("%s\t%d\t%s\t%s %s\t%s\t%s\t%0.0f\t%0.1f\t%s\t%s\n",
       $1, $2, $3, $5,$6, width,
       $6, width*$7*0.01, 100.0*2*$6/width, $7, $9);
}' > cpgIsland.bed

    # verify longest unique chrom name:
    cut -f1 cpgIsland.bed | awk '{print length($0)}' | sort -rn | head -1
    #	20
    # update the length 14 in the template to be 16:
    sed -e "s/14/20/" $HOME/kent/src/hg/lib/cpgIslandExt.sql > cpgIslandExt.sql

    cd /hive/data/genomes/mm10/bed/cpgIsland
    hgLoadBed mm10 cpgIslandExt -tab -sqlTable=cpgIslandExt.sql cpgIsland.bed
# Loaded 16023 elements of size 10

    featureBits mm10 cpgIslandExt
    #	10495450 bases of 2652783500 (0.396%) in intersection
    # compare to previous:
    featureBits mm9 cpgIslandExt
    #	10496250 bases of 2620346127 (0.401%) in intersection

    # there should be no output from checkTableCoords:
    checkTableCoords -verboseBlocks -table=cpgIslandExt mm10

    #	cleanup, unless you want to move them to the genscan procedure below
    rm -fr hardMaskedFa

#########################################################################
# GENSCAN GENE PREDICTIONS (DONE - 2012-02-09,10 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/genscan
    cd /hive/data/genomes/mm10/bed/genscan
    # use a previously existing genscan binary
    ln -s ../../../mm9/bed/genscan/hg3rdParty .

    # create hard masked .fa files
    mkdir -p hardMaskedFa
    cut -f1 ../../chrom.sizes | while read C
do
    echo ${C}
    twoBitToFa ../../mm10.2bit:$C stdout \
	| maskOutFa stdin hard hardMaskedFa/${C}.fa
done

    # Generate a list file, genome.list, of all the hard-masked contig chunks:
    find ./hardMaskedFa/ -type f | sed -e 's#^./##' > genome.list

    wc -l genome.list
    #	66 genome.list

    # Run on small cluster (more mem than big cluster).
    ssh encodek
    cd /hive/data/genomes/mm10/bed/genscan
    # Make 3 subdirectories for genscan to put their output files in
    mkdir gtf pep subopt
    # Create template file, template, for gensub2.  For example (3-line file):
    cat << '_EOF_' > template
#LOOP
/cluster/bin/x86_64/gsBig {check in exists+ $(path1)} {check out exists gtf/$(root1).gtf} -trans={check out exists pep/$(root1).pep} -subopt={check out exists subopt/$(root1).bed} -exe=hg3rdParty/genscanlinux/genscan -par=hg3rdParty/genscanlinux/HumanIso.smat -tmp=/tmp -window=2400000
#ENDLOOP
'_EOF_'
    # << emacs
    gensub2 genome.list single template jobList
    para create jobList
    para try
    para check ... etc...
    para time
# Crashed: 2 jobs
# CPU time in finished jobs:     171336s    2855.60m    47.59h    1.98d  0.005 y
# IO & Wait Time:                   261s       4.35m     0.07h    0.00d  0.000 y
# Average job time:                2640s      44.00m     0.73h    0.03d
# Longest finished job:           22618s     376.97m     6.28h    0.26d
# Submission to last job:         28682s     478.03m     7.97h    0.33d

    # one of the two crashed jobs was just a stray line in the jobList,
    # somehow a line with the string: '_EOF_' got in there.

    # as with mm9, chr7 did not work.  Break it up into pieces
    mkdir /hive/data/genomes/mm10/bed/genscan/chr7Split
    cd /hive/data/genomes/mm10/bed/genscan/chr7Split
    grep chr7 ../../../jkStuff/mm10.nonBridged.lift | grep -v random \
	> chr7.nonBridged.lift
    faToTwoBit ../hardMaskedFa/chr7.fa chr7.2bit
    ~/kent/src/hg/utils/lft2BitToFa.pl chr7.2bit chr7.nonBridged.lift \
	| sed -e "s/chr7./chr7_/" > chr7.nonBridged.fa
    faSplit sequence chr7.nonBridged.fa 100 split7/chr7_

    ln -s ../../../../mm9/bed/genscan/hg3rdParty .
    echo '#!/bin/sh' > cmdList.sh
    ls split7 | while read F
do
echo "/cluster/bin/x86_64/gsBig split7/${F} gtf/${F}.gtf} -trans=pep/${F}.pep} -subopt=subopt/${F}.bed -exe=hg3rdParty/genscanlinux/genscan -par=hg3rdParty/genscanlinux/HumanIso.smat -tmp=/tmp -window=2400000 &"
done >> cmdList.sh
    echo "wait" >> cmdList.sh
    chmod +x cmdList.sh
    mkdir gtf pep subopt
    time ./cmdList.sh > run.log 2>&1
    # about 20 minutes

    # fix the names in the lift file
    cat chr7.nonBridged.lift | sed -e "s/chr7./chr7_/" > chr7.lift

    # the sed mangling will provide unique names for them all, but they
    #	will not be in the strict numerical order that genscan usually produces
    cat gtf/chr7_*.gtf | liftUp -type=.gtf stdout chr7.lift error stdin \
	| sed -e "s/chr7_0\([0-4]\)\./chr7.\1/g" > chr7.gtf
    cat subopt/chr7_*.bed | liftUp -type=.bed stdout chr7.lift error stdin \
	| sed -e "s/chr7_0\([0-4]\)\./chr7.\1/g" > chr7.subopt.bed
    cat pep/chr7_*.pep | sed -e "s/chr7_0\([0-4]\)\./chr7.\1/g" > chr7.pep
    cp -p chr7.pep ../pep
    cp -p chr7.gtf ../gtf
    cp -p chr7.subopt.bed ../subopt/chr7.bed

    find ./gtf -type f | xargs -n 256 endsInLf -zeroOk

    # Concatenate results:
    cd /hive/data/genomes/mm10/bed/genscan
    find ./gtf -type f | xargs cat > genscan.gtf
    find ./pep -type f | xargs cat > genscan.pep
    find ./subopt -type f | xargs cat > genscanSubopt.bed

    # Load into the data/genomesbase (without -genePredExt because no frame info):
    # Don't load the Pep anymore -- redundant since it's from genomic.
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/genscan
    # to construct a local file with the genePred business:
    gtfToGenePred genscan.gtf genscan.gp
    # this produces exactly the same thing and loads the table:
    ldHgGene -gtf mm10 genscan genscan.gtf
    #	Read 45012 transcripts in 323529 lines in 1 files
    #	45012 groups 59 seqs 1 sources 1 feature types
    #	45012 gene predictions
    hgLoadBed mm10 genscanSubopt genscanSubopt.bed
    #	Read 526572 elements of size 6 from genscanSubopt.bed
    featureBits mm10 genscan
    #	55743040 bases of 2652783500 (2.101%) in intersection
    # previously:
    featureBits mm9 genscan
    #	55293837 bases of 2620346127 (2.110%) in intersection

#########################################################################
# CREATE MICROSAT TRACK (DONE - 2012-02-09 - Hiram
     ssh hgwdev
     mkdir /cluster/data/genomes/mm10/bed/microsat
     cd /cluster/data/genomes/mm10/bed/microsat
     awk '($5==2 || $5==3) && $6 >= 15 && $8 == 100 && $9 == 0 {printf("%s\t%s\t%s\t%dx%s\n", $1, $2, $3, $6, $16);}' \
	../simpleRepeat/simpleRepeat.bed > microsat.bed
    hgLoadBed mm10 microsat microsat.bed
    #	Read 197237 elements of size 4 from microsat.bed

#########################################################################
#  BLATSERVERS ENTRY (DONE - 2012-02-09 - Hiram)
#	After getting a blat server assigned by the Blat Server Gods,
    ssh hgwdev

    hgsql -e 'INSERT INTO blatServers (db, host, port, isTrans, canPcr) \
	VALUES ("mm10", "blat13", "17832", "1", "0"); \
	INSERT INTO blatServers (db, host, port, isTrans, canPcr) \
	VALUES ("mm10", "blat13", "17833", "0", "1");' \
	    hgcentraltest
    #	test it with some sequence

############################################################################
# set default position the same as was mm9 via blat
#	(DONE - 2012-02-09 - Hiram)
    hgsql -e \
'update dbDb set defaultPos="chr12:56694976-56714605" where name="mm10";' \
	hgcentraltest

############################################################################
# constructing downloads (DONE - 2012-02-09 - Hiram)
    cd /hive/data/genomes/mm10
    # some of the smaller bits are missing the simple repeat results
    time makeDownloads.pl -allowMissedTrfs -workhorse=hgwdev mm10
    #	real    41m42.408s

    # edit the README files in goldenPath/*/README.txt

#########################################################################
# create pushQ entry (DONE - 2012-02-09 - Hiram)
    # first make sure all.joiner is up to date and has this new organism
    # a keys check should be clean:
    cd ~/kent/src/hg/makeDb/schema
    joinerCheck -data/genomesbase=mm10 -keys all.joiner

    mkdir /hive/data/genomes/mm10/pushQ
    cd /hive/data/genomes/mm10/pushQ
    makePushQSql.pl mm10 > mm10.sql 2> stderr.out
    # check stderr.out for no significant problems, it is common to see:
# WARNING: hgwdev does not have /gbdb/mm10/wib/gc5Base.wib
# WARNING: hgwdev does not have /gbdb/mm10/wib/quality.wib
# WARNING: hgwdev does not have /gbdb/mm10/bbi/quality.bw
# WARNING: mm10 does not have seq
# WARNING: mm10 does not have extFile
# *** All done!

    # which are not real problem
    # if some tables are not identified:
# WARNING: Could not tell (from trackDb, all.joiner and hardcoded lists of
# supporting and genbank tables) which tracks to assign these tables to:
#  list of tables will be in the output
# put them in manually after loading the pushQ entry
    scp -p mm10.sql hgwbeta:/tmp
    ssh hgwbeta
    cd /tmp
    hgsql qapushq < mm10.sql

#########################################################################
# lifting ensGene track from mm9 (DONE - 2012-02-22 - Hiram)
    # no gene tracks yet on mm10.  liftUp mm9 ensGenes to mm10
    # history of mm9 ensGene indicates it is the same as v64 release
    #	with v65 being identical
    mkdir /hive/data/genomes/mm10/bed/ensGene
    cd /hive/data/genomes/mm10/bed/ensGene
    hgsql -N -e "select * from ensGene;" mm9 | cut -f2- > mm9.ensGene.gp
    liftOver -genePred mm9.ensGene.gp \
	/gbdb/mm9/liftOver/mm9ToMm10.over.chain.gz \
	mm10.lifted.ensGene.gp unmapped.ensGene.gp
    wc -l *.gp
    #	95651 mm10.lifted.ensGene.gp
    #	95883 mm9.ensGene.gp
    #	464 unmapped.ensGene.gp

    hgLoadGenePred -skipInvalid -genePredExt mm10 ensGene mm10.lifted.ensGene.gp
    #	Warning: skipping 118 invalid genePreds

    # make a list of what did get loaded:
    hgsql -N -e "select name from ensGene;" mm10 \
	| sort -u > mm10.name.ensGene.txt
    wc -l mm10.name.ensGene.txt
    #	95533 mm10.name.ensGene.txt

    hgsql -N -e "select * from ensPep;" mm9 | sort > mm9.ensPep.tab
    hgsql -N -e "select * from ensGtp;" mm9 | sort -k2,2 > mm9.ensGtp.tab
    hgsql -N -e "select * from ensemblToGeneName;" mm9 | sort -k1,1 \
	> mm9.ensemblToGeneName.tab
    hgsql -N -e "select * from ensemblSource;" mm9 | sort -k1,1 \
	> mm9.ensemblSource.tab

    # select out ensGtp records that match with the names in mm10 ensGene:
    join -1 2 -2 1 -o "1.1,1.2,1.3" mm9.ensGtp.tab mm10.name.ensGene.txt \
	| tr '[ ]' '[\t]' > mm10.ensGtp.tab
    wc -l *.ensGtp.tab
    #	95533 mm10.ensGtp.tab
    #	95883 mm9.ensGtp.tab

    # select out ensPep records that match with the names in mm10 ensGene:
    join -1 1 -2 2 -o "1.1,1.2" mm9.ensPep.tab mm10.ensGtp.tab \
	| tr '[ ]' '[\t]' > mm10.ensPep.tab

    wc -l mm9.ensPep.tab mm10.ensPep.tab
    #	55798 mm9.ensPep.tab
    #	55485 mm10.ensPep.tab

    # select out ensemblSource records that match the mm10 ensGene names:
    join -1 1 -2 1 -o "1.1,1.2" mm9.ensemblSource.tab mm10.name.ensGene.txt \
	| tr '[ ]' '[\t]' > mm10.ensemblSource.tab
    wc -l mm9.ensemblSource.tab mm10.ensemblSource.tab
  95883 mm9.ensemblSource.tab
  95533 mm10.ensemblSource.tab

    # select out ensemblToGeneName records that match the mm10 ensGene names:
    join -1 1 -2 1 -o "1.1,1.2" mm9.ensemblToGeneName.tab \
	mm10.name.ensGene.txt | tr '[ ]' '[\t]' > mm10.ensemblToGeneName.tab
    wc -l mm9.ensemblToGeneName.tab mm10.ensemblToGeneName.tab
    #	95883 mm9.ensemblToGeneName.tab
    #	95533 mm10.ensemblToGeneName.tab

    hgPepPred mm10 tab ensPep mm10.ensPep.tab
    hgLoadSqlTab mm10 ensGtp ~/kent/src/hg/lib/ensGtp.sql mm10.ensGtp.tab
    sed -e "s/15/18/" ~/kent/src/hg/lib/ensemblSource.sql > ensemblSource.sql
    hgLoadSqlTab mm10 ensemblSource ensemblSource.sql mm10.ensemblSource.tab

    # find sizes for indexes
  NL=`awk '{print length($1)}' mm10.ensemblToGeneName.tab | sort -rn | head -1`
  VL=`awk '{print length($2)}' mm10.ensemblToGeneName.tab | sort -rn | head -1`
    # construct sql definition with appropriate index sizes
    sed -e "s/ knownTo / ensemblToGeneName /; s/known gene/ensGen/; s/INDEX(name(12)/PRIMARY KEY(name($NL)/; s/value(12)/value($VL)/" \
	~/kent/src/hg/lib/knownTo.sql > ensemblToGeneName.sql

    hgLoadSqlTab mm10 ensemblToGeneName ensemblToGeneName.sql \
	mm10.ensemblToGeneName.tab

hgsql -e 'INSERT INTO trackVersion \
    (db, name, who, version, updateTime, comment, source, dateReference) \
    VALUES("mm10", "ensGene", "hiram", "65", now(), \
        "lifted from mm9 ensGene 65", \
        "lifted from mm9 ensGene 65", \
        "dec2011" );' hgFixed

#########################################################################
# Swap lastz Human hg19 (DONE - 2012-03-08 - Hiram)
    # original alignment to hg19
    cd /hive/data/genomes/hg19/bed/lastzMm10.2012-03-07
    cat fb.hg19.chainMm10Link.txt
    #	1021265143 bases of 2897316137 (35.249%) in intersection

    #	and the swap
    mkdir /hive/data/genomes/mm10/bed/blastz.hg19.swap
    cd /hive/data/genomes/mm10/bed/blastz.hg19.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/hg19/bed/lastzMm10.2012-03-07/DEF \
	-swap -noLoadChainSplit -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    72m32.794s
    cat fb.mm10.chainHg19Link.txt
    #	1014045890 bases of 2652783500 (38.226%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s blastz.hg19.swap lastz.hg19

#########################################################################
# LASTZ RAT Rn4 (DONE - 2012-03-08 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzRn4.2012-03-08
    cd /hive/data/genomes/mm10/bed/lastzRn4.2012-03-08

    cat << '_EOF_' > DEF
# mouse vs rat
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# Specially tuned blastz parameters from Webb Miller
BLASTZ_O=600
BLASTZ_E=150
BLASTZ_Y=15000
BLASTZ_T=2
BLASTZ_K=4500
BLASTZ_Q=/scratch/data/blastz/human_chimp.v2.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Rat Rn4
SEQ2_DIR=/scratch/data/rn4/rn4.2bit
SEQ2_LEN=/scratch/data/rn4/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0

BASE=/hive/data/genomes/mm10/bed/lastzRn4.2012-03-08
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S rn4Mm10
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-bigClusterHub=swarm -chainMinScore=5000 -chainLinearGap=medium \
	-noLoadChainSplit -syntenicNet -workhorse=hgwdev \
	-smallClusterHub=encodek > do.log 2>&1 &
    #	real    129m48.444s
    cat fb.mm10.chainRn4Link.txt
    #	1449612208 bases of 2652783500 (54.645%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzRn4.2012-03-08 lastz.rn4

    #	and the swap
    mkdir /hive/data/genomes/rn4/bed/blastz.mm10.swap
    cd /hive/data/genomes/rn4/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzRn4.2012-03-08/DEF \
	-swap -bigClusterHub=swarm -chainMinScore=5000 -chainLinearGap=medium \
	-noLoadChainSplit -syntenicNet -workhorse=hgwdev \
	-smallClusterHub=encodek > swap.log 2>&1 &
    #	real    71m10.645s
    cat fb.rn4.chainMm10Link.txt
    #	1449012636 bases of 2571531505 (56.348%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/rn4/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# LASTZ Gorilla gorGor3 (DONE - 2012-03-08 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzGorGor3.2012-03-08
    cd /hive/data/genomes/mm10/bed/lastzGorGor3.2012-03-08

    cat << '_EOF_' > DEF
# gorilla vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Gorilla GorGor3
SEQ2_DIR=/scratch/data/gorGor3/gorGor3.2bit
SEQ2_LEN=/scratch/data/gorGor3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=300

BASE=/hive/data/genomes/mm10/bed/lastzGorGor3.2012-03-08
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10GorGor3
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    625m17.180s
    cat fb.mm10.chainGorGor3Link.txt
    #	901610588 bases of 2652783500 (33.987%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzGorGor3.2012-03-08 lastz.gorGor3

    mkdir /hive/data/genomes/gorGor3/bed/blastz.mm10.swap
    cd /hive/data/genomes/gorGor3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzGorGor3.2012-03-08/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    91m3.616s
    cat fb.gorGor3.chainMm10Link.txt
    #	969595533 bases of 2822760080 (34.349%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/gorGor3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ Gibbon nomLeu1 (DONE - 2012-03-08 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzNomLeu1.2012-03-08
    cd /hive/data/genomes/mm10/bed/lastzNomLeu1.2012-03-08

    cat << '_EOF_' > DEF
# gibbon vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Gibbon NomLeu1
SEQ2_DIR=/scratch/data/nomLeu1/nomLeu1.2bit
SEQ2_LEN=/scratch/data/nomLeu1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=300

BASE=/hive/data/genomes/mm10/bed/lastzNomLeu1.2012-03-08
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10NomLeu1
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    556m26.589s
    cat fb.mm10.chainNomLeu1Link.txt
    #	905455766 bases of 2652783500 (34.132%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzNomLeu1.2012-03-08 lastz.nomLeu1

    mkdir /hive/data/genomes/nomLeu1/bed/blastz.mm10.swap
    cd /hive/data/genomes/nomLeu1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzNomLeu1.2012-03-08/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    66m50.839s
    cat fb.nomLeu1.chainMm10Link.txt
    #	892362811 bases of 2756591777 (32.372%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/nomLeu1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ Rhesus rheMac3 (DONE - 2012-03-08 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzRheMac3.2012-03-08
    cd /hive/data/genomes/mm10/bed/lastzRheMac3.2012-03-08

    cat << '_EOF_' > DEF
# rhesus vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Rhesus RheMac3
SEQ2_DIR=/scratch/data/rheMac3/rheMac3.2bit
SEQ2_LEN=/scratch/data/rheMac3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=300

BASE=/hive/data/genomes/mm10/bed/lastzRheMac3.2012-03-08
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10RheMac3
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    596m55.622s
    cat fb.mm10.chainRheMac3Link.txt
    #	900117108 bases of 2652783500 (33.931%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzRheMac3.2012-03-08 lastz.rheMac3

    mkdir /hive/data/genomes/rheMac3/bed/blastz.mm10.swap
    cd /hive/data/genomes/rheMac3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzRheMac3.2012-03-08/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    69m5.839s
    cat fb.rheMac3.chainMm10Link.txt
    #	883164992 bases of 2639145830 (33.464%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/rheMac3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ Baboon papHam1 (DONE - 2012-03-09 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzPapHam1.2012-03-09
    cd /hive/data/genomes/mm10/bed/lastzPapHam1.2012-03-09

    cat << '_EOF_' > DEF
# baboon vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Baboon PapHam1
SEQ2_DIR=/scratch/data/papHam1/papHam1.2bit
SEQ2_LEN=/scratch/data/papHam1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=500

BASE=/hive/data/genomes/mm10/bed/lastzPapHam1.2012-03-09
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10PapHam1
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1138m52.716s
    cat fb.mm10.chainPapHam1Link.txt
    #	890718423 bases of 2652783500 (33.577%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzPapHam1.2012-03-09 lastz.papHam1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzPapHam1.2012-03-09
    time doRecipBest.pl mm10 papHam1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    899m48.908s

    mkdir /hive/data/genomes/papHam1/bed/blastz.mm10.swap
    cd /hive/data/genomes/papHam1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzPapHam1.2012-03-09/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    548m15.438s
    cat fb.mm10.chainPapHam1Link.txt
    #	878016290 bases of 2741867288 (32.023%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/papHam1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# Swap ponAbe2 lastz (DONE - 2012-03-09 - Hiram)
    # original alignment result:
    cd /hive/data/genomes/ponAbe2/bed/lastzMm10.2012-03-08
    cat fb.ponAbe2.chainMm10Link.txt
    #	946932454 bases of 3093572278 (30.610%) in intersection

    #	and the swap
    mkdir /hive/data/genomes/mm10/bed/blastz.ponAbe2.swap
    cd /hive/data/genomes/mm10/bed/blastz.ponAbe2.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/ponAbe2/bed/lastzMm10.2012-03-08/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    72m38.550s
    cat fb.mm10.chainPonAbe2Link.txt
    #	915093866 bases of 2652783500 (34.496%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s blastz.ponAbe2.swap lastz.ponAbe2

##############################################################################
# LASTZ Squirrel monkey saiBol1 (DONE - 2012-03-09 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzSaiBol1.2012-03-09
    cd /hive/data/genomes/mm10/bed/lastzSaiBol1.2012-03-09

    cat << '_EOF_' > DEF
# squirrel monkey vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Squirrel monkey SaiBol1
SEQ2_DIR=/hive/data/genomes/saiBol1/saiBol1.2bit
SEQ2_LEN=/hive/data/genomes/saiBol1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzSaiBol1.2012-03-09
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10SaiBol1
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    538m42.643s
    cat fb.mm10.chainSaiBol1Link.txt
    #	857872391 bases of 2652783500 (32.339%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzSaiBol1.2012-03-09 lastz.saiBol1

    mkdir /hive/data/genomes/saiBol1/bed/blastz.mm10.swap
    cd /hive/data/genomes/saiBol1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzSaiBol1.2012-03-09/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    59m36.306s
    cat fb.saiBol1.chainMm10Link.txt
    #	838457857 bases of 2477131095 (33.848%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/saiBol1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ Marmoset calJac3 (DONE - 2012-03-09 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzCalJac3.2012-03-09
    cd /hive/data/genomes/mm10/bed/lastzCalJac3.2012-03-09

    cat << '_EOF_' > DEF
# marmoset vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Marmoset monkey CalJac3
SEQ2_DIR=/scratch/data/calJac3/calJac3.2bit
SEQ2_LEN=/scratch/data/calJac3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=100

BASE=/hive/data/genomes/mm10/bed/lastzCalJac3.2012-03-09
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10CalJac3
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    529m39.657s
    cat fb.mm10.chainCalJac3Link.txt
    #	860830771 bases of 2652783500 (32.450%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzCalJac3.2012-03-09 lastz.calJac3

    mkdir /hive/data/genomes/calJac3/bed/blastz.mm10.swap
    cd /hive/data/genomes/calJac3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzCalJac3.2012-03-09/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    67m21.635s
    cat fb.calJac3.chainMm10Link.txt
    #	861565545 bases of 2752505800 (31.301%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/calJac3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ Chimp PanTro4 (DONE - 2012-03-09 - Hiram)
    mkdir /hive/data/genomes/panTro3/bed/lastzMm10.2012-03-09
    cd /hive/data/genomes/panTro3/bed/lastzMm10.2012-03-09

    cat << '_EOF_' > DEF
# chimp vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=10000000
SEQ1_LAP=10000

# QUERY: Chimp PanTro4
SEQ2_DIR=/hive/data/genomes/panTro4/panTro4.2bit
SEQ2_LEN=/hive/data/genomes/panTro4/chrom.sizes
SEQ2_CHUNK=20000000
SEQ2_LAP=0
SEQ2_LIMIT=200

BASE=/hive/data/genomes/mm10/bed/lastzPanTro4.2012-03-09
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10PanTro4
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet -noLoadChainSplit \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    682m53.046s
    cat fb.mm10.chainPanTro4Link.txt
    #	919836299 bases of 2652783500 (34.674%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzPanTro4.2012-03-09 lastz.panTro4

    mkdir /hive/data/genomes/panTro4/bed/blastz.mm10.swap
    cd /hive/data/genomes/panTro4/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzPanTro4.2012-03-09/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    73m23.855s
    cat fb.panTro4.chainMm10Link.txt
    #	926540065 bases of 2902338967 (31.924%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/panTro4/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ tarsier tarSyr1 (DONE - 2012-03-10 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzTarSyr1.2012-03-10
    cd /hive/data/genomes/mm10/bed/lastzTarSyr1.2012-03-10

    cat << '_EOF_' > DEF
# tarsier vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: tarsier TarSyr1
SEQ2_DIR=/scratch/data/tarSyr1/tarSyr1.2bit
SEQ2_LEN=/scratch/data/tarSyr1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=800

BASE=/hive/data/genomes/mm10/bed/lastzTarSyr1.2012-03-10
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10TarSyr1
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2457m45.759s
    cat fb.mm10.chainTarSyr1Link.txt
    #	651517559 bases of 2652783500 (24.560%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzTarSyr1.2012-03-10 lastz.tarSyr1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzTarSyr1.2012-03-10
    time doRecipBest.pl mm10 tarSyr1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1176m19.336s

    mkdir /hive/data/genomes/tarSyr1/bed/blastz.mm10.swap
    cd /hive/data/genomes/tarSyr1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzTarSyr1.2012-03-10/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    746m30.852s
    cat fb.tarSyr1.chainMm10Link.txt
    #	691746721 bases of 2768536343 (24.986%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/tarSyr1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# Swap chimp panTro3 to Mm10 (DONE - 2012-03-12 - Hiram)
    # original alignment on panTro3
    cd /hive/data/genomes/panTro3/bed/lastzMm10.2012-03-08
    cat fb.panTro3.chainMm10Link.txt
    #	929073028 bases of 2900529764 (32.031%) in intersection

    # and this swap:
    mkdir /hive/data/genomes/mm10/bed/blastz.panTro3.swap
    cd /hive/data/genomes/mm10/bed/blastz.panTro3.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/panTro3/bed/lastzMm10.2012-03-08/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    68m46.408s
    cat fb.mm10.chainPanTro3Link.txt
    #	922491113 bases of 2652783500 (34.774%) in intersection

##############################################################################
# LASTZ bushbaby otoGar3 (DONE - 2012-03-13 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzOtoGar3.2012-03-13
    cd /hive/data/genomes/mm10/bed/lastzOtoGar3.2012-03-13

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# bushbaby vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: bushbaby OtoGar3
SEQ2_DIR=/hive/data/genomes/otoGar3/otoGar3.2bit
SEQ2_LEN=/hive/data/genomes/otoGar3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzOtoGar3.2012-03-13
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10OtoGar3
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    757m32.438s
    cat fb.mm10.chainOtoGar3Link.txt
    #	790408953 bases of 2652783500 (29.795%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOtoGar3.2012-03-13 lastz.otoGar3

    mkdir /hive/data/genomes/otoGar3/bed/blastz.mm10.swap
    cd /hive/data/genomes/otoGar3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOtoGar3.2012-03-13/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    61m18.952s
    cat fb.otoGar3.chainMm10Link.txt
    #	776907989 bases of 2359530453 (32.926%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/otoGar3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ mouse lemur micMur1 (DONE - 2012-03-13 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzMicMur1.2012-03-13
    cd /hive/data/genomes/mm10/bed/lastzMicMur1.2012-03-13

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# mouse lemur vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: mouse lemur MicMur1
SEQ2_DIR=/scratch/data/micMur1/micMur1.2bit
SEQ2_LEN=/scratch/data/micMur1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=400

BASE=/hive/data/genomes/mm10/bed/lastzMicMur1.2012-03-13
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10MicMur1
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    687m41.863s
    cat fb.mm10.chainMicMur1Link.txt
    #	706607444 bases of 2652783500 (26.636%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzMicMur1.2012-03-13 lastz.micMur1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzMicMur1.2012-03-13
    time doRecipBest.pl mm10 micMur1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    877m18.105s

    mkdir /hive/data/genomes/micMur1/bed/blastz.mm10.swap
    cd /hive/data/genomes/micMur1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzMicMur1.2012-03-13/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    116m54.411s
    cat fb.micMur1.chainMm10Link.txt
    #	696025630 bases of 1852394361 (37.574%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/micMur1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ squirrel speTri2 (DONE - 2012-03-15 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/lastzSpeTri2.2012-03-15
    cd /hive/data/genomes/mm10/bed/lastzSpeTri2.2012-03-15

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# squirrel vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: squirrel SpeTri2
SEQ2_DIR=/hive/data/genomes/speTri2/speTri2.2bit
SEQ2_LEN=/hive/data/genomes/speTri2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzSpeTri2.2012-03-15
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10SpeTri2
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    935m27.893s
    cat fb.mm10.chainSpeTri2Link.txt
    #	907715417 bases of 2652783500 (34.217%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzSpeTri2.2012-03-15 lastz.speTri2

    mkdir /hive/data/genomes/speTri2/bed/blastz.mm10.swap
    cd /hive/data/genomes/speTri2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzSpeTri2.2012-03-15/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    74m41.819s
    #	real    116m54.411s
    cat fb.speTri2.chainMm10Link.txt
    #	906956512 bases of 2311060300 (39.244%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/speTri2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ kangaroo rat dipOrd1 (DONE - 2012-03-15 - Hiram)
    #	establish a screen to control this job
    screen -S mm10DipOrd1
    mkdir /hive/data/genomes/mm10/bed/lastzDipOrd1.2012-03-15
    cd /hive/data/genomes/mm10/bed/lastzDipOrd1.2012-03-15

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# kangaroo rat vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: kangaroo rat DipOrd1
SEQ2_DIR=/scratch/data/dipOrd1/dipOrd1.2bit
SEQ2_LEN=/scratch/data/dipOrd1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=400

BASE=/hive/data/genomes/mm10/bed/lastzDipOrd1.2012-03-15
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    867m19.972s
    cat fb.mm10.chainDipOrd1Link.txt
    #	516232678 bases of 2652783500 (19.460%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzDipOrd1.2012-03-15 lastz.dipOrd1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzDipOrd1.2012-03-15
    time doRecipBest.pl mm10 dipOrd1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    914m20.405s

    mkdir /hive/data/genomes/dipOrd1/bed/blastz.mm10.swap
    cd /hive/data/genomes/dipOrd1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzDipOrd1.2012-03-15/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    115m1.497s
    cat fb.dipOrd1.chainMm10Link.txt
    #	507580668 bases of 1844961421 (27.512%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/dipOrd1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ Naked mole-rat hetGla1 (DONE - 2012-03-15 - Hiram)
    #	establish a screen to control this job
    screen -S mm10HetGla1
    mkdir /hive/data/genomes/mm10/bed/lastzHetGla1.2012-03-15
    cd /hive/data/genomes/mm10/bed/lastzHetGla1.2012-03-15

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# Naked mole-rat vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Naked mole-rat HetGla1
SEQ2_DIR=/scratch/data/hetGla1/hetGla1.2bit
SEQ2_LEN=/scratch/data/hetGla1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=200

BASE=/hive/data/genomes/mm10/bed/lastzHetGla1.2012-03-15
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    745m15.097s
    cat fb.mm10.chainHetGla1Link.txt
    #	853221843 bases of 2652783500 (32.163%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzHetGla1.2012-03-15 lastz.hetGla1

    mkdir /hive/data/genomes/hetGla1/bed/blastz.mm10.swap
    cd /hive/data/genomes/hetGla1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzHetGla1.2012-03-15/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    74m26.471s
    cat fb.hetGla1.chainMm10Link.txt
    #	885195861 bases of 2430064805 (36.427%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/hetGla1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ horse equCab2 (DONE - 2012-03-16 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10EquCab2
    mkdir /hive/data/genomes/mm10/bed/lastzEquCab2.2012-03-16
    cd /hive/data/genomes/mm10/bed/lastzEquCab2.2012-03-16

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# horse vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: horse EquCab2
SEQ2_DIR=/scratch/data/equCab2/equCab2.2bit
SEQ2_LEN=/scratch/data/equCab2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=1

BASE=/hive/data/genomes/mm10/bed/lastzEquCab2.2012-03-16
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    566m34.024s
    cat fb.mm10.chainEquCab2Link.txt
    #	912967841 bases of 2652783500 (34.415%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzEquCab2.2012-03-16 lastz.equCab2

    mkdir /hive/data/genomes/equCab2/bed/blastz.mm10.swap
    cd /hive/data/genomes/equCab2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzEquCab2.2012-03-16/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    87m2.261s
    cat fb.equCab2.chainMm10Link.txt
    #	901995882 bases of 2428790173 (37.138%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/equCab2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ guinea pig cavPor3 (DONE - 2012-03-16 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10CavPor3
    mkdir /hive/data/genomes/mm10/bed/lastzCavPor3.2012-03-16
    cd /hive/data/genomes/mm10/bed/lastzCavPor3.2012-03-16

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# guinea pig vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: guinea pig CavPor3
SEQ2_DIR=/scratch/data/cavPor3/cavPor3.2bit
SEQ2_LEN=/scratch/data/cavPor3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=10

BASE=/hive/data/genomes/mm10/bed/lastzCavPor3.2012-03-16
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1523m35.729s
    cat fb.mm10.chainCavPor3Link.txt
    #	754642254 bases of 2652783500 (28.447%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzCavPor3.2012-03-16 lastz.cavPor3

    mkdir /hive/data/genomes/cavPor3/bed/blastz.mm10.swap
    cd /hive/data/genomes/cavPor3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzCavPor3.2012-03-16/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    80m23.870s
    cat fb.cavPor3.chainMm10Link.txt
    #	775452752 bases of 2663369733 (29.115%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/cavPor3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ alpaca vicPac1 (DONE - 2012-03-16 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10VicPac1
    mkdir /hive/data/genomes/mm10/bed/lastzVicPac1.2012-03-16
    cd /hive/data/genomes/mm10/bed/lastzVicPac1.2012-03-16

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# guinea pig vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: alpaca VicPac1
SEQ2_DIR=/scratch/data/vicPac1/vicPac1.2bit
SEQ2_LEN=/scratch/data/vicPac1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=700

BASE=/hive/data/genomes/mm10/bed/lastzVicPac1.2012-03-16
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2049m38.674s
    cat fb.mm10.chainVicPac1Link.txt
    #	600477253 bases of 2652783500 (22.636%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzVicPac1.2012-03-16 lastz.vicPac1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzVicPac1.2012-03-16
    time doRecipBest.pl mm10 vicPac1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    824m37.107s

    mkdir /hive/data/genomes/vicPac1/bed/blastz.mm10.swap
    cd /hive/data/genomes/vicPac1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzVicPac1.2012-03-16/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    159m21.952s
    cat fb.vicPac1.chainMm10Link.txt
    #	610885692 bases of 1922910435 (31.769%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/vicPac1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ dolphin turTru1 (DONE - 2012-03-16 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10TurTru1
    mkdir /hive/data/genomes/mm10/bed/lastzTurTru1.2012-03-16
    cd /hive/data/genomes/mm10/bed/lastzTurTru1.2012-03-16

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# dolphin vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: dolphin TurTru1
SEQ2_DIR=/scratch/data/turTru1/turTru1.2bit
SEQ2_LEN=/scratch/data/turTru1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=300

BASE=/hive/data/genomes/mm10/bed/lastzTurTru1.2012-03-16
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1484m14.609s
    cat fb.mm10.chainTurTru1Link.txt
    #	762961671 bases of 2652783500 (28.761%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzTurTru1.2012-03-16 lastz.turTru1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzTurTru1.2012-03-16
    time doRecipBest.pl mm10 turTru1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    733m37.272s

    mkdir /hive/data/genomes/turTru1/bed/blastz.mm10.swap
    cd /hive/data/genomes/turTru1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzTurTru1.2012-03-16/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    79m38.703s
    cat fb.turTru1.chainMm10Link.txt
    #	744359707 bases of 2298444090 (32.385%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/turTru1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ tree shrew tupBel1 (DONE - 2012-03-16 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10TupBel1
    mkdir /hive/data/genomes/mm10/bed/lastzTupBel1.2012-03-16
    cd /hive/data/genomes/mm10/bed/lastzTupBel1.2012-03-16

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# tree shrew vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: tree shrew TupBel1
SEQ2_DIR=/scratch/data/tupBel1/tupBel1.2bit
SEQ2_LEN=/scratch/data/tupBel1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=400

BASE=/hive/data/genomes/mm10/bed/lastzTupBel1.2012-03-16
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1731m30.449s
    cat fb.mm10.chainTupBel1Link.txt
    #	524337666 bases of 2652783500 (19.766%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzTupBel1.2012-03-16 lastz.tupBel1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzTupBel1.2012-03-16
    time doRecipBest.pl mm10 tupBel1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1090m30.429s

    mkdir /hive/data/genomes/tupBel1/bed/blastz.mm10.swap
    cd /hive/data/genomes/tupBel1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzTupBel1.2012-03-16/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    136m7.163s
    cat fb.tupBel1.chainMm10Link.txt
    #	537379661 bases of 2137225476 (25.144%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/tupBel1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ pig susScr2 (DONE - 2012-03-16 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10SusScr2
    mkdir /hive/data/genomes/mm10/bed/lastzSusScr2.2012-03-16
    cd /hive/data/genomes/mm10/bed/lastzSusScr2.2012-03-16

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# pig vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: pig SusScr2
SEQ2_DIR=/scratch/data/susScr2/susScr2.2bit
SEQ2_LEN=/scratch/data/susScr2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=1

BASE=/hive/data/genomes/mm10/bed/lastzSusScr2.2012-03-16
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1272m57.727s
    cat fb.mm10.chainSusScr2Link.txt
    #	616716602 bases of 2652783500 (23.248%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzSusScr2.2012-03-16 lastz.susScr2

    mkdir /hive/data/genomes/susScr2/bed/blastz.mm10.swap
    cd /hive/data/genomes/susScr2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzSusScr2.2012-03-16/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    62m47.465s
    cat fb.susScr2.chainMm10Link.txt
    #	656498040 bases of 2231298548 (29.422%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/susScr2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ rabbit oryCun2 (DONE - 2012-03-16 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10OryCun2
    mkdir /hive/data/genomes/mm10/bed/lastzOryCun2.2012-03-16
    cd /hive/data/genomes/mm10/bed/lastzOryCun2.2012-03-16

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# rabbit vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: rabbit OryCun2
SEQ2_DIR=/scratch/data/oryCun2/oryCun2.2bit
SEQ2_LEN=/scratch/data/oryCun2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=20

BASE=/hive/data/genomes/mm10/bed/lastzOryCun2.2012-03-16
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1412m58.641s
    cat fb.mm10.chainOryCun2Link.txt
    #	669778489 bases of 2652783500 (25.248%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOryCun2.2012-03-16 lastz.oryCun2

    mkdir /hive/data/genomes/oryCun2/bed/blastz.mm10.swap
    cd /hive/data/genomes/oryCun2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOryCun2.2012-03-16/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    64m40.959s
    cat fb.oryCun2.chainMm10Link.txt
    #	668643668 bases of 2604023284 (25.677%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/oryCun2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ sloth choHof1 (DONE - 2012-03-19 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10ChoHof1
    mkdir /hive/data/genomes/mm10/bed/lastzChoHof1.2012-03-19
    cd /hive/data/genomes/mm10/bed/lastzChoHof1.2012-03-19

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# sloth vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: sloth ChoHof1
SEQ2_DIR=/scratch/data/choHof1/choHof1.2bit
SEQ2_LEN=/scratch/data/choHof1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=800

BASE=/hive/data/genomes/mm10/bed/lastzChoHof1.2012-03-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    # rebooted hgwdev during first swarm run, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-continue=cat -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > cat.log 2>&1 &
    #	Elapsed time: 65m26s
    cat fb.mm10.chainChoHof1Link.txt
    #	477994856 bases of 2652783500 (18.019%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzChoHof1.2012-03-19 lastz.choHof1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzChoHof1.2012-03-19
    time doRecipBest.pl mm10 choHof1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1171m56.481s

    mkdir /hive/data/genomes/choHof1/bed/blastz.mm10.swap
    cd /hive/data/genomes/choHof1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzChoHof1.2012-03-19/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    1613m3.348s
    cat fb.choHof1.chainMm10Link.txt
    #	488047499 bases of 2060419685 (23.687%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/choHof1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ megabat pteVam1 (DONE - 2012-03-19 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10PteVam1
    mkdir /hive/data/genomes/mm10/bed/lastzPteVam1.2012-03-19
    cd /hive/data/genomes/mm10/bed/lastzPteVam1.2012-03-19

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# megabat vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: megabat PteVam1
SEQ2_DIR=/scratch/data/pteVam1/pteVam1.2bit
SEQ2_LEN=/scratch/data/pteVam1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=200

BASE=/hive/data/genomes/mm10/bed/lastzPteVam1.2012-03-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1843m33.186s
    cat fb.mm10.chainPteVam1Link.txt
    #	725414059 bases of 2652783500 (27.345%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzPteVam1.2012-03-19 lastz.pteVam1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzPteVam1.2012-03-19
    time doRecipBest.pl mm10 pteVam1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    743m57.901s

    mkdir /hive/data/genomes/pteVam1/bed/blastz.mm10.swap
    cd /hive/data/genomes/pteVam1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzPteVam1.2012-03-19/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	Elapsed time: 75m35s
    cat fb.pteVam1.chainMm10Link.txt
    #	710519911 bases of 1839436660 (38.627%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/pteVam1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ elephant loxAfr3 (DONE - 2012-03-19 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10LoxAfr3
    mkdir /hive/data/genomes/mm10/bed/lastzLoxAfr3.2012-03-19
    cd /hive/data/genomes/mm10/bed/lastzLoxAfr3.2012-03-19

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# elephant vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: elephant LoxAfr3
SEQ2_DIR=/scratch/data/loxAfr3/loxAfr3.2bit
SEQ2_LEN=/scratch/data/loxAfr3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzLoxAfr3.2012-03-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1848m11.111s
    cat fb.mm10.chainLoxAfr3Link.txt
    #	685029753 bases of 2652783500 (25.823%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzLoxAfr3.2012-03-19 lastz.loxAfr3

    mkdir /hive/data/genomes/loxAfr3/bed/blastz.mm10.swap
    cd /hive/data/genomes/loxAfr3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzLoxAfr3.2012-03-19/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	Elapsed time: 73m14s
    cat fb.loxAfr3.chainMm10Link.txt
    #	674108752 bases of 3118565340 (21.616%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/loxAfr3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ cat felCat4 (DONE - 2012-03-19 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10FelCat4
    mkdir /hive/data/genomes/mm10/bed/lastzFelCat4.2012-03-19
    cd /hive/data/genomes/mm10/bed/lastzFelCat4.2012-03-19

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# cat vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: cat FelCat4
SEQ2_DIR=/scratch/data/felCat4/felCat4.2bit
SEQ2_LEN=/scratch/data/felCat4/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=300

BASE=/hive/data/genomes/mm10/bed/lastzFelCat4.2012-03-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2010m48.963s
    cat fb.mm10.chainFelCat4Link.txt
    #	637531191 bases of 2652783500 (24.033%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzFelCat4.2012-03-19 lastz.felCat4

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzFelCat4.2012-03-19
    time doRecipBest.pl mm10 felCat4 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1135m12.207s

    mkdir /hive/data/genomes/felCat4/bed/blastz.mm10.swap
    cd /hive/data/genomes/felCat4/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzFelCat4.2012-03-19/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	 Elapsed time: 88m12s
    cat fb.felCat4.chainMm10Link.txt
    #	616167655 bases of 1990635005 (30.953%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/felCat4/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ panda ailMel1 (DONE - 2012-03-19 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10AilMel1
    mkdir /hive/data/genomes/mm10/bed/lastzAilMel1.2012-03-19
    cd /hive/data/genomes/mm10/bed/lastzAilMel1.2012-03-19

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# panda vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: panda AilMel1
SEQ2_DIR=/scratch/data/ailMel1/ailMel1.2bit
SEQ2_LEN=/scratch/data/ailMel1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=200

BASE=/hive/data/genomes/mm10/bed/lastzAilMel1.2012-03-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # forgot to copy to the log
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium
    #	real    1914m15.921s
    cat fb.mm10.chainAilMel1Link.txt
    #	821806974 bases of 2652783500 (30.979%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzAilMel1.2012-03-19 lastz.ailMel1

    mkdir /hive/data/genomes/ailMel1/bed/blastz.mm10.swap
    cd /hive/data/genomes/ailMel1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzAilMel1.2012-03-19/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	Elapsed time: 65m50s
    cat fb.ailMel1.chainMm10Link.txt
    #	798482731 bases of 2245312831 (35.562%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/ailMel1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ dog canFam3 (DONE - 2012-03-19 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10CanFam3
    mkdir /hive/data/genomes/mm10/bed/lastzCanFam3.2012-03-19
    cd /hive/data/genomes/mm10/bed/lastzCanFam3.2012-03-19

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# dog vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: dog CanFam3
SEQ2_DIR=/hive/data/genomes/canFam3/canFam3.2bit
SEQ2_LEN=/hive/data/genomes/canFam3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=20

BASE=/hive/data/genomes/mm10/bed/lastzCanFam3.2012-03-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # forgot to copy to the log
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1883m21.850s
    cat fb.mm10.chainCanFam3Link.txt
    #	773114990 bases of 2652783500 (29.144%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzCanFam3.2012-03-19 lastz.canFam3

    mkdir /hive/data/genomes/canFam3/bed/blastz.mm10.swap
    cd /hive/data/genomes/canFam3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzCanFam3.2012-03-19/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	Elapsed time: 63m22s
    cat fb.canFam3.chainMm10Link.txt
    #	756678903 bases of 2392715236 (31.624%) in intersectio
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/canFam3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ armadillo dasNov2 (DONE - 2012-03-21 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10DasNov2
    mkdir /hive/data/genomes/mm10/bed/lastzDasNov2.2012-03-21
    cd /hive/data/genomes/mm10/bed/lastzDasNov2.2012-03-21

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# armadillo vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: armadillo DasNov2
SEQ2_DIR=/scratch/data/dasNov2/dasNov2.2bit
SEQ2_LEN=/scratch/data/dasNov2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=800

BASE=/hive/data/genomes/mm10/bed/lastzDasNov2.2012-03-21
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2655m49.904s
    cat fb.mm10.chainDasNov2Link.txt
    #	451070039 bases of 2652783500 (17.004%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzDasNov2.2012-03-21 lastz.dasNov2

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzDasNov2.2012-03-21
    time doRecipBest.pl mm10 dasNov2 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1163m1.023s

    mkdir /hive/data/genomes/dasNov2/bed/blastz.mm10.swap
    cd /hive/data/genomes/dasNov2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzDasNov2.2012-03-21/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    201m9.701s
    cat fb.dasNov2.chainMm10Link.txt
    #	461142417 bases of 2371493872 (19.445%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/dasNov2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ microbat myoLuc2 (DONE - 2012-03-21 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10MyoLuc2
    mkdir /hive/data/genomes/mm10/bed/lastzMyoLuc2.2012-03-21
    cd /hive/data/genomes/mm10/bed/lastzMyoLuc2.2012-03-21

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# microbat vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: microbat MyoLuc2
SEQ2_DIR=/scratch/data/myoLuc2/myoLuc2.2bit
SEQ2_LEN=/scratch/data/myoLuc2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzMyoLuc2.2012-03-21
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1033m38.184s
    cat fb.mm10.chainMyoLuc2Link.txt
    #	646292112 bases of 2652783500 (24.363%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzMyoLuc2.2012-03-21 lastz.myoLuc2

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzMyoLuc2.2012-03-21
    time doRecipBest.pl mm10 myoLuc2 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #   real    29m16.249s

    mkdir /hive/data/genomes/myoLuc2/bed/blastz.mm10.swap
    cd /hive/data/genomes/myoLuc2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzMyoLuc2.2012-03-21/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    54m5.607s
    cat fb.myoLuc2.chainMm10Link.txt
    #	661704053 bases of 1966419868 (33.650%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/myoLuc2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ cow bosTau7 (DONE - 2012-03-21 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10BosTau7
    mkdir /hive/data/genomes/mm10/bed/lastzBosTau7.2012-03-21
    cd /hive/data/genomes/mm10/bed/lastzBosTau7.2012-03-21

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# cow vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: cow BosTau7
SEQ2_DIR=/scratch/data/bosTau7/bosTau7.2bit
SEQ2_LEN=/scratch/data/bosTau7/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzBosTau7.2012-03-21
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1151m20.445s
    cat fb.mm10.chainBosTau7Link.txt
    #	696498363 bases of 2652783500 (26.255%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzBosTau7.2012-03-21 lastz.bosTau7

    mkdir /hive/data/genomes/bosTau7/bed/blastz.mm10.swap
    cd /hive/data/genomes/bosTau7/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzBosTau7.2012-03-21/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    77m58.759s
    cat fb.bosTau7.chainMm10Link.txt
    #	711923052 bases of 2804673174 (25.383%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/bosTau7/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ sheep oviAri1 (DONE - 2012-03-21 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10OviAri1
    mkdir /hive/data/genomes/mm10/bed/lastzOviAri1.2012-03-21
    cd /hive/data/genomes/mm10/bed/lastzOviAri1.2012-03-21

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# sheep vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: sheep OviAri1
SEQ2_DIR=/scratch/data/oviAri1/oviAri1.2bit
SEQ2_LEN=/scratch/data/oviAri1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzOviAri1.2012-03-21
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    892m33.068s
    cat fb.mm10.chainOviAri1Link.txt
    #	406955832 bases of 2652783500 (15.341%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOviAri1.2012-03-21 lastz.oviAri1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzOviAri1.2012-03-21
    time doRecipBest.pl mm10 oviAri1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1183m43.488s

    mkdir /hive/data/genomes/oviAri1/bed/blastz.mm10.swap
    cd /hive/data/genomes/oviAri1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOviAri1.2012-03-21/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    30m5.554s
    cat fb.oviAri1.chainMm10Link.txt
    #	383499897 bases of 1201271277 (31.925%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/oviAri1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ rock hyrax proCap1 (DONE - 2012-03-21 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10ProCap1
    mkdir /hive/data/genomes/mm10/bed/lastzProCap1.2012-03-21
    cd /hive/data/genomes/mm10/bed/lastzProCap1.2012-03-21

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# rock hyrax vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: rock hyrax ProCap1
SEQ2_DIR=/scratch/data/proCap1/proCap1.2bit
SEQ2_LEN=/scratch/data/proCap1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=600

BASE=/hive/data/genomes/mm10/bed/lastzProCap1.2012-03-21
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2859m51.317s
    cat fb.mm10.chainProCap1Link.txt
    #	401804601 bases of 2652783500 (15.147%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzProCap1.2012-03-21 lastz.proCap1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzProCap1.2012-03-21
    time doRecipBest.pl mm10 proCap1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1083m57.139s

    mkdir /hive/data/genomes/proCap1/bed/blastz.mm10.swap
    cd /hive/data/genomes/proCap1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzProCap1.2012-03-21/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    165m10.285s
    cat fb.proCap1.chainMm10Link.txt
    #	390409777 bases of 2407847681 (16.214%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/proCap1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ pika ochPri2 (DONE - 2012-03-22 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10OchPri2
    mkdir /hive/data/genomes/mm10/bed/lastzOchPri2.2012-03-22
    cd /hive/data/genomes/mm10/bed/lastzOchPri2.2012-03-22

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# pika vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: pika OchPri2
SEQ2_DIR=/scratch/data/ochPri2/ochPri2.2bit
SEQ2_LEN=/scratch/data/ochPri2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=500

BASE=/hive/data/genomes/mm10/bed/lastzOchPri2.2012-03-22
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2578m43.648s
    cat fb.mm10.chainOchPri2Link.txt
    #	385766335 bases of 2652783500 (14.542%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOchPri2.2012-03-22 lastz.ochPri2

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzOchPri2.2012-03-22
    time doRecipBest.pl mm10 ochPri2 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1036m29.080s

    mkdir /hive/data/genomes/ochPri2/bed/blastz.mm10.swap
    cd /hive/data/genomes/ochPri2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOchPri2.2012-03-22/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    103m34.369s
    cat fb.ochPri2.chainMm10Link.txt
    #	382959642 bases of 1923624051 (19.908%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/ochPri2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ hedgehog eriEur1 (DONE - 2012-03-22 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10EriEur1
    mkdir /hive/data/genomes/mm10/bed/lastzEriEur1.2012-03-22
    cd /hive/data/genomes/mm10/bed/lastzEriEur1.2012-03-22

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# hedgehog vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: hedgehog EriEur1
SEQ2_DIR=/scratch/data/eriEur1/eriEur1.2bit
SEQ2_LEN=/scratch/data/eriEur1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=700

BASE=/hive/data/genomes/mm10/bed/lastzEriEur1.2012-03-22
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    3006m41.470s
    cat fb.mm10.chainEriEur1Link.txt
    #	261447061 bases of 2652783500 (9.856%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzEriEur1.2012-03-22 lastz.eriEur1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzEriEur1.2012-03-22
    time doRecipBest.pl mm10 eriEur1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1171m41.349s

    mkdir /hive/data/genomes/eriEur1/bed/blastz.mm10.swap
    cd /hive/data/genomes/eriEur1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzEriEur1.2012-03-22/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    241m24.183s
    cat fb.eriEur1.chainMm10Link.txt
    #	261605017 bases of 2133134836 (12.264%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/eriEur1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ tenrec echTel1 (DONE - 2012-03-22 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10EchTel1
    mkdir /hive/data/genomes/mm10/bed/lastzEchTel1.2012-03-22
    cd /hive/data/genomes/mm10/bed/lastzEchTel1.2012-03-22

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# tenrec vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: tenrec EchTel1
SEQ2_DIR=/scratch/data/echTel1/echTel1.2bit
SEQ2_LEN=/scratch/data/echTel1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=700

BASE=/hive/data/genomes/mm10/bed/lastzEchTel1.2012-03-22
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    3047m28.723s
    cat fb.mm10.chainEchTel1Link.txt
    #	290413150 bases of 2652783500 (10.947%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzEchTel1.2012-03-22 lastz.echTel1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzEchTel1.2012-03-22
    time doRecipBest.pl mm10 echTel1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1201m39.275s

    mkdir /hive/data/genomes/echTel1/bed/blastz.mm10.swap
    cd /hive/data/genomes/echTel1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzEchTel1.2012-03-22/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    269m52.619s
    cat fb.echTel1.chainMm10Link.txt
    #	298082139 bases of 2111581369 (14.117%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/echTel1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ shrew sorAra1 (DONE - 2012-03-22 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10SorAra1
    mkdir /hive/data/genomes/mm10/bed/lastzSorAra1.2012-03-22
    cd /hive/data/genomes/mm10/bed/lastzSorAra1.2012-03-22

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# shrew vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: shrew SorAra1
SEQ2_DIR=/scratch/data/sorAra1/sorAra1.2bit
SEQ2_LEN=/scratch/data/sorAra1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=500

BASE=/hive/data/genomes/mm10/bed/lastzSorAra1.2012-03-22
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2600m22.528s
    cat fb.mm10.chainSorAra1Link.txt
    #	248874412 bases of 2652783500 (9.382%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzSorAra1.2012-03-22 lastz.sorAra1

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzSorAra1.2012-03-22
    time doRecipBest.pl mm10 sorAra1 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1074m22.651s

    mkdir /hive/data/genomes/sorAra1/bed/blastz.mm10.swap
    cd /hive/data/genomes/sorAra1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzSorAra1.2012-03-22/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    141m38.806s
    cat fb.sorAra1.chainMm10Link.txt
    #	248692550 bases of 1832864697 (13.569%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/sorAra1/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ wallaby macEug2 (DONE - 2012-03-22 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10MacEug2
    mkdir /hive/data/genomes/mm10/bed/lastzMacEug2.2012-03-22
    cd /hive/data/genomes/mm10/bed/lastzMacEug2.2012-03-22

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# wallaby vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: wallaby MacEug2
SEQ2_DIR=/scratch/data/macEug2/macEug2.2bit
SEQ2_LEN=/scratch/data/macEug2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=500

BASE=/hive/data/genomes/mm10/bed/lastzMacEug2.2012-03-22
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    2893m50.341s
    cat fb.mm10.chainMacEug2Link.txt
    #	115481931 bases of 2652783500 (4.353%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzMacEug2.2012-03-22 lastz.macEug2

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzMacEug2.2012-03-22
    time doRecipBest.pl mm10 macEug2 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    1032m58.798s

    mkdir /hive/data/genomes/macEug2/bed/blastz.mm10.swap
    cd /hive/data/genomes/macEug2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzMacEug2.2012-03-22/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    130m7.404s
    cat fb.macEug2.chainMm10Link.txt
    #	112811810 bases of 2536076957 (4.448%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/macEug2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ RAT Rn5 (DONE - 2012-03-23 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10Rn5
    mkdir /hive/data/genomes/mm10/bed/lastzRn5.2012-03-23
    cd /hive/data/genomes/mm10/bed/lastzRn5.2012-03-23

    cat << '_EOF_' > DEF
# mouse vs rat
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# From tuning experiment between mouse chr12:15000000-25000000 and
#       rat chr6:38000000-48000000
BLASTZ_O=600
BLASTZ_E=55
BLASTZ_Y=5000
BLASTZ_T=2
BLASTZ_K=3000
BLASTZ_L=3000
BLASTZ_Q=/hive/data/genomes/mm10/bed/lastzRn5.2012-03-23/mouse_rat_2.q

BLASTZ_ABRIDGE_REPEATS=1

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_SMSK=/scratch/data/mm10/notInRat
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Rat Rn5
SEQ2_DIR=/hive/data/genomes/rn5/rn5.2bit
SEQ2_LEN=/hive/data/genomes/rn5/chrom.sizes
SEQ2_SMSK=/hive/data/genomes/rn5/bed/linSpecRep/notInMouse
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=10

BASE=/hive/data/genomes/mm10/bed/lastzRn5.2012-03-23
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S rn5Mm10
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-bigClusterHub=swarm -chainMinScore=5000 -chainLinearGap=medium \
	-noLoadChainSplit -syntenicNet -workhorse=hgwdev \
	-smallClusterHub=encodek > do.log 2>&1 &
    # broken lastz run when SMSK files did not exist for some of the
    #	Rn5 contigs - made empty files for those and completed, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	-continue=cat `pwd`/DEF \
	-bigClusterHub=swarm -chainMinScore=5000 -chainLinearGap=medium \
	-noLoadChainSplit -syntenicNet -workhorse=hgwdev \
	-smallClusterHub=encodek > cat.log 2>&1 &
    #	real    285m28.458s
    cat fb.mm10.chainRn5Link.txt
    #	1786721927 bases of 2652783500 (67.353%) in intersection
    # FYI: rn4 was:
    #	1449612208 bases of 2652783500 (54.645%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzRn5.2012-03-23 lastz.rn5

    #	and the swap
    mkdir /hive/data/genomes/rn5/bed/blastz.mm10.swap
    cd /hive/data/genomes/rn5/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzRn5.2012-03-23/DEF \
	-swap -bigClusterHub=swarm -chainMinScore=5000 -chainLinearGap=medium \
	-noLoadChainSplit -syntenicNet -workhorse=hgwdev \
	-smallClusterHub=encodek > swap.log 2>&1 &
    #	real    121m21.029s
    cat fb.rn5.chainMm10Link.txt
    #	1808154679 bases of 2572853723 (70.278%) in intersection
    # FYI, rn4 was:
    #	1449012636 bases of 2571531505 (56.348%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/rn5/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# LASTZ Manatee triMan1 (DONE - 2012-03-29 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10TriMan1
    mkdir /hive/data/genomes/mm10/bed/lastzTriMan1.2012-03-29
    cd /hive/data/genomes/mm10/bed/lastzTriMan1.2012-03-29

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# manatee vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: manatee TriMan1
SEQ2_DIR=/hive/data/genomes/triMan1/triMan1.2bit
SEQ2_LEN=/hive/data/genomes/triMan1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzTriMan1.2012-03-29
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #	real    1455m24.772s
    cat fb.mm10.chainTriMan1Link.txt
    #	704207702 bases of 2652783500 (26.546%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzTriMan1.2012-03-29 lastz.triMan1

    mkdir /hive/data/genomes/triMan1/bed/blastz.mm10.swap
    cd /hive/data/genomes/triMan1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzTriMan1.2012-03-29/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    62m33.530s
    cat fb.triMan1.chainMm10Link.txt
    #	682557025 bases of 2769099677 (24.649%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/triMan1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz Opossum monDom5 (DONE - 2012-03-29 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10MonDom5
    mkdir /hive/data/genomes/mm10/bed/lastzMonDom5.2012-03-29
    cd /hive/data/genomes/mm10/bed/lastzMonDom5.2012-03-29

    cat << '_EOF_' > DEF
# Mouse vs. opossum
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Opossum monDom5
SEQ2_DIR=/scratch/data/monDom5/monDom5.2bit
SEQ2_LEN=/scratch/data/monDom5/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0

BASE=/hive/data/genomes/mm10/bed/lastzMonDom5.2012-03-29
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    #	Can't do this when there are only the single small set of chroms
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    1792m40.071s

    cat fb.mm10.chainMonDom5Link.txt
    #	254245903 bases of 2652783500 (9.584%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzMonDom5.2012-03-29 lastz.monDom5

    #	and for the swap
    mkdir /hive/data/genomes/monDom5/bed/blastz.mm10.swap
    cd /hive/data/genomes/monDom5/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzMonDom5.2012-03-29/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    73m49.230s
    cat  fb.monDom5.chainMm10Link.txt
    #	252291401 bases of 3501660299 (7.205%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/monDom5/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz Tasmanian Devil sarHar1 (DONE - 2012-03-29 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10SarHar1
    mkdir /hive/data/genomes/mm10/bed/lastzSarHar1.2012-03-29
    cd /hive/data/genomes/mm10/bed/lastzSarHar1.2012-03-29

    cat << '_EOF_' > DEF
# Mouse vs. tasmanian devil
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: tasmanian devil sarHar1
SEQ2_DIR=/scratch/data/sarHar1/sarHar1.2bit
SEQ2_LEN=/scratch/data/sarHar1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0

BASE=/hive/data/genomes/mm10/bed/lastzSarHar1.2012-03-29
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    1208m55.866s

    cat fb.mm10.chainSarHar1Link.txt
    #	224935746 bases of 2652783500 (8.479%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzSarHar1.2012-03-29 lastz.sarHar1

    #	and for the swap
    mkdir /hive/data/genomes/sarHar1/bed/blastz.mm10.swap
    cd /hive/data/genomes/sarHar1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzSarHar1.2012-03-29/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    45m53.015s
    cat  fb.sarHar1.chainMm10Link.txt
    #	231249436 bases of 2931539702 (7.888%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/sarHar1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz budgerigar melUnd1 (DONE - 2012-03-29 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10MelUnd1
    mkdir /hive/data/genomes/mm10/bed/lastzMelUnd1.2012-03-29
    cd /hive/data/genomes/mm10/bed/lastzMelUnd1.2012-03-29

    cat << '_EOF_' > DEF
# Mouse vs. budgerigar
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: budgerigar melUnd1
SEQ2_DIR=/hive/data/genomes/melUnd1/melUnd1.2bit
SEQ2_LEN=/hive/data/genomes/melUnd1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzMelUnd1.2012-03-29
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    883m58.198s

    cat fb.mm10.chainMelUnd1Link.txt
    #	95217653 bases of 2652783500 (3.589%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzMelUnd1.2012-03-29 lastz.melUnd1

    #	and for the swap
    mkdir /hive/data/genomes/melUnd1/bed/blastz.mm10.swap
    cd /hive/data/genomes/melUnd1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzMelUnd1.2012-03-29/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    9m9.260s
    cat  fb.melUnd1.chainMm10Link.txt
    #	79867911 bases of 1086614815 (7.350%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/melUnd1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz platypus ornAna1 (DONE - 2012-03-29 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10OrnAna1
    mkdir /hive/data/genomes/mm10/bed/lastzOrnAna1.2012-03-29
    cd /hive/data/genomes/mm10/bed/lastzOrnAna1.2012-03-29

    cat << '_EOF_' > DEF
# Mouse vs. platypus
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: platypus ornAna1
SEQ2_DIR=/scratch/data/ornAna1/ornAna1.2bit
SEQ2_LEN=/scratch/data/ornAna1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=400

BASE=/hive/data/genomes/mm10/bed/lastzOrnAna1.2012-03-29
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    1264m1.056s

    cat fb.mm10.chainOrnAna1Link.txt
    #	141873792 bases of 2652783500 (5.348%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOrnAna1.2012-03-29 lastz.ornAna1

    #	and for the swap
    mkdir /hive/data/genomes/ornAna1/bed/blastz.mm10.swap
    cd /hive/data/genomes/ornAna1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOrnAna1.2012-03-29/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    49m45.308s
    cat  fb.ornAna1.chainMm10Link.txt
    #	135101083 bases of 1842236818 (7.334%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/ornAna1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz turtle chrPic1 (DONE - 2012-03-29 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10ChrPic1
    mkdir /hive/data/genomes/mm10/bed/lastzChrPic1.2012-03-29
    cd /hive/data/genomes/mm10/bed/lastzChrPic1.2012-03-29

    cat << '_EOF_' > DEF
# Mouse vs. turtle
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: turtle chrPic1
SEQ2_DIR=/hive/data/genomes/chrPic1/chrPic1.2bit
SEQ2_LEN=/hive/data/genomes/chrPic1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=200

BASE=/hive/data/genomes/mm10/bed/lastzChrPic1.2012-03-29
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    1243m2.518s
    cat fb.mm10.chainChrPic1Link.txt
    #	125499965 bases of 2652783500 (4.731%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzChrPic1.2012-03-29 lastz.chrPic1

    #	and for the swap
    mkdir /hive/data/genomes/chrPic1/bed/blastz.mm10.swap
    cd /hive/data/genomes/chrPic1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzChrPic1.2012-03-29/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    19m26.835s
    cat  fb.chrPic1.chainMm10Link.txt
    #	118436838 bases of 2158289746 (5.488%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/chrPic1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz chicken galGal4 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10GalGal4
    mkdir /hive/data/genomes/mm10/bed/lastzGalGal4.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzGalGal4.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. chicken
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: chicken galGal4
SEQ2_DIR=/hive/data/genomes/galGal4/galGal4.2bit
SEQ2_LEN=/hive/data/genomes/galGal4/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzGalGal4.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    109m21.068s
    #	broken swarm cluster, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    57m24.155s
    cat fb.mm10.chainGalGal4Link.txt
    #	97510773 bases of 2652783500 (3.676%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzGalGal4.2012-04-02 lastz.galGal4

    #	and for the swap
    mkdir /hive/data/genomes/galGal4/bed/blastz.mm10.swap
    cd /hive/data/genomes/galGal4/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzGalGal4.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    95m50.996s
    cat  fb.galGal4.chainMm10Link.txt
    #	83660034 bases of 1032854810 (8.100%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/galGal4/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz zebra finch taeGut1 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10TaeGut1
    mkdir /hive/data/genomes/mm10/bed/lastzTaeGut1.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzTaeGut1.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. zebra finch
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: zebra finch taeGut1
SEQ2_DIR=/scratch/data/taeGut1/taeGut1.2bit
SEQ2_LEN=/scratch/data/taeGut1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=5

BASE=/hive/data/genomes/mm10/bed/lastzTaeGut1.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    106m11.612s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    29m11.090s
    cat fb.mm10.chainTaeGut1Link.txt
    #	95469341 bases of 2652783500 (3.599%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzTaeGut1.2012-04-02 lastz.taeGut1

    #	and for the swap
    mkdir /hive/data/genomes/taeGut1/bed/blastz.mm10.swap
    cd /hive/data/genomes/taeGut1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzTaeGut1.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    37m17.483s
    cat  fb.taeGut1.chainMm10Link.txt
    #	89312133 bases of 1222864691 (7.304%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/taeGut1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz lizard anoCar2 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10AnoCar2
    mkdir /hive/data/genomes/mm10/bed/lastzAnoCar2.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzAnoCar2.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. lizard
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: lizard anoCar2
SEQ2_DIR=/scratch/data/anoCar2/anoCar2.2bit
SEQ2_LEN=/scratch/data/anoCar2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=15

BASE=/hive/data/genomes/mm10/bed/lastzAnoCar2.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    103m17.133s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    43m2.183s
    cat fb.mm10.chainAnoCar2Link.txt
    #	88356459 bases of 2652783500 (3.331%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzAnoCar2.2012-04-02 lastz.anoCar2

    #	and for the swap
    mkdir /hive/data/genomes/anoCar2/bed/blastz.mm10.swap
    cd /hive/data/genomes/anoCar2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzAnoCar2.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    97m50.599s
    cat  fb.anoCar2.chainMm10Link.txt
    #	84865552 bases of 1701353770 (4.988%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/anoCar2/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz turkey melGal1 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10MelGal1
    mkdir /hive/data/genomes/mm10/bed/lastzMelGal1.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzMelGal1.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. turkey
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: turkey melGal1
SEQ2_DIR=/scratch/data/melGal1/melGal1.2bit
SEQ2_LEN=/scratch/data/melGal1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=15

BASE=/hive/data/genomes/mm10/bed/lastzMelGal1.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    101m17.902s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    20m47.771s
    cat fb.mm10.chainMelGal1Link.txt
    #	93132953 bases of 2652783500 (3.511%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzMelGal1.2012-04-02 lastz.melGal1

    #	and for the swap
    mkdir /hive/data/genomes/melGal1/bed/blastz.mm10.swap
    cd /hive/data/genomes/melGal1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzMelGal1.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    88m39.591s
    cat  fb.melGal1.chainMm10Link.txt
    #	76848161 bases of 935922386 (8.211%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/melGal1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz frog xenTro3 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10XenTro3
    mkdir /hive/data/genomes/mm10/bed/lastzXenTro3.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzXenTro3.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. frog
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: frog xenTro3
SEQ2_DIR=/scratch/data/xenTro3/xenTro3.2bit
SEQ2_LEN=/scratch/data/xenTro3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=40

BASE=/hive/data/genomes/mm10/bed/lastzXenTro3.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    99m10.611s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    37m52.678s
    cat fb.mm10.chainXenTro3Link.txt
    #	82900338 bases of 2652783500 (3.125%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzXenTro3.2012-04-02 lastz.xenTro3

    #	and for the swap
    mkdir /hive/data/genomes/xenTro3/bed/blastz.mm10.swap
    cd /hive/data/genomes/xenTro3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzXenTro3.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    53m19.485s
    cat  fb.xenTro3.chainMm10Link.txt
    #	90345130 bases of 1358334882 (6.651%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/xenTro3/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz coelacanth latCha1 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10LatCha1
    mkdir /hive/data/genomes/mm10/bed/lastzLatCha1.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzLatCha1.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. coelacanth
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: coelacanth latCha1
SEQ2_DIR=/hive/data/genomes/latCha1/latCha1.2bit
SEQ2_LEN=/hive/data/genomes/latCha1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzLatCha1.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    95m34.477s
    #	broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    214m7.324s
    cat fb.mm10.chainLatCha1Link.txt
    #	72036116 bases of 2652783500 (2.715%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzLatCha1.2012-04-02 lastz.latCha1

    #	and for the swap
    mkdir /hive/data/genomes/latCha1/bed/blastz.mm10.swap
    cd /hive/data/genomes/latCha1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzLatCha1.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    14m44.600s
    cat  fb.latCha1.chainMm10Link.txt
    #	73798131 bases of 2183592768 (3.380%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/latCha1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz atlantic cod gadMor1 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10GadMor1
    mkdir /hive/data/genomes/mm10/bed/lastzGadMor1.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzGadMor1.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. atlantic cod
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: atlantic cod gadMor1
SEQ2_DIR=/hive/data/genomes/gadMor1/gadMor1.2bit
SEQ2_LEN=/hive/data/genomes/gadMor1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=700

BASE=/hive/data/genomes/mm10/bed/lastzGadMor1.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    91m23.642s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    39m41.194s
    cat fb.mm10.chainGadMor1Link.txt
    #	45795692 bases of 2652783500 (1.726%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzGadMor1.2012-04-02 lastz.gadMor1

    #	and for the swap
    mkdir /hive/data/genomes/gadMor1/bed/blastz.mm10.swap
    cd /hive/data/genomes/gadMor1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzGadMor1.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    62m58.963s
    cat  fb.gadMor1.chainMm10Link.txt
    #	41406507 bases of 608038597 (6.810%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/gadMor1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz nile tilapia oreNil1 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10OreNil1
    mkdir /hive/data/genomes/mm10/bed/lastzOreNil1.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzOreNil1.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. nile tilapia
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: nile tilapia oreNil1
SEQ2_DIR=/scratch/data/oreNil1/oreNil1.2bit
SEQ2_LEN=/scratch/data/oreNil1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=10

BASE=/hive/data/genomes/mm10/bed/lastzOreNil1.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    89m6.727s
    #	broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    24m3.960s
    cat fb.mm10.chainOreNil1Link.txt
    #	51915568 bases of 2652783500 (1.957%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOreNil1.2012-04-02 lastz.oreNil1

    #	and for the swap
    mkdir /hive/data/genomes/oreNil1/bed/blastz.mm10.swap
    cd /hive/data/genomes/oreNil1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOreNil1.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    90m55.298s
    cat  fb.oreNil1.chainMm10Link.txt
    #	49709461 bases of 816084674 (6.091%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/oreNil1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz stickleback gasAcu1 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10GasAcu1
    mkdir /hive/data/genomes/mm10/bed/lastzGasAcu1.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzGasAcu1.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. stickleback
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: stickleback gasAcu1
SEQ2_DIR=/scratch/data/gasAcu1/gasAcu1.2bit
SEQ2_LEN=/scratch/data/gasAcu1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=1

BASE=/hive/data/genomes/mm10/bed/lastzGasAcu1.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    87m5.963s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    9m49.199s
    cat fb.mm10.chainGasAcu1Link.txt
    #	53469711 bases of 2652783500 (2.016%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzGasAcu1.2012-04-02 lastz.gasAcu1

    #	and for the swap
    mkdir /hive/data/genomes/gasAcu1/bed/blastz.mm10.swap
    cd /hive/data/genomes/gasAcu1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzGasAcu1.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    12m58.072s
    cat  fb.gasAcu1.chainMm10Link.txt
    #	48802831 bases of 446627861 (10.927%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/gasAcu1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz fugu fr3 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10Fr3
    mkdir /hive/data/genomes/mm10/bed/lastzFr3.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzFr3.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. fugu
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: fugu fr3
SEQ2_DIR=/scratch/data/fr3/fr3.2bit
SEQ2_LEN=/scratch/data/fr3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzFr3.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    84m37.070s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    171m16.627s
    cat fb.mm10.chainFr3Link.txt
    #	47460021 bases of 2652783500 (1.789%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzFr3.2012-04-02 lastz.fr3

    #	and for the swap
    mkdir /hive/data/genomes/fr3/bed/blastz.mm10.swap
    cd /hive/data/genomes/fr3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzFr3.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    7m13.151s
    cat  fb.fr3.chainMm10Link.txt
    #	42586058 bases of 350961831 (12.134%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/fr3/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz tetraodon tetNig2 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10TetNig2
    mkdir /hive/data/genomes/mm10/bed/lastzTetNig2.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzTetNig2.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. tetraodon
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: tetraodon tetNig2
SEQ2_DIR=/scratch/data/tetNig2/tetNig2.2bit
SEQ2_LEN=/scratch/data/tetNig2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzTetNig2.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    13m21.638s
    cat fb.mm10.chainTetNig2Link.txt
    #	46035322 bases of 2652783500 (1.735%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzTetNig2.2012-04-02 lastz.tetNig2

    #	and for the swap
    mkdir /hive/data/genomes/tetNig2/bed/blastz.mm10.swap
    cd /hive/data/genomes/tetNig2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzTetNig2.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    7m24.115s
    cat  fb.tetNig2.chainMm10Link.txt
    #	41242926 bases of 302314788 (13.642%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/tetNig2/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz zebrafish danRer7 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10DanRer7
    mkdir /hive/data/genomes/mm10/bed/lastzDanRer7.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzDanRer7.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. zebrafish
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: zebrafish danRer7
SEQ2_DIR=/scratch/data/danRer7/danRer7.2bit
SEQ2_LEN=/scratch/data/danRer7/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzDanRer7.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    80m32.118s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    40m27.762s
    cat fb.mm10.chainDanRer7Link.txt
    #	69028912 bases of 2652783500 (2.602%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzDanRer7.2012-04-02 lastz.danRer7

    #	and for the swap
    mkdir /hive/data/genomes/danRer7/bed/blastz.mm10.swap
    cd /hive/data/genomes/danRer7/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzDanRer7.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    109m49.939s
    cat  fb.danRer7.chainMm10Link.txt
    #	72001768 bases of 1409770109 (5.107%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/danRer7/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz medaka oryLat2 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10OryLat2
    mkdir /hive/data/genomes/mm10/bed/lastzOryLat2.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzOryLat2.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. medaka
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: medaka oryLat2
SEQ2_DIR=/scratch/data/oryLat2/oryLat2.2bit
SEQ2_LEN=/scratch/data/oryLat2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzOryLat2.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    78m53.408s
    #	broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    #	real    113m29.462s
    cat fb.mm10.chainOryLat2Link.txt
    #	51344841 bases of 2652783500 (1.936%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOryLat2.2012-04-02 lastz.oryLat2

    #	and for the swap
    mkdir /hive/data/genomes/oryLat2/bed/blastz.mm10.swap
    cd /hive/data/genomes/oryLat2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOryLat2.2012-04-02/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #	real    7m52.846s
    cat  fb.oryLat2.chainMm10Link.txt
    #	45954178 bases of 700386597 (6.561%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/oryLat2/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz lamprey petMar1 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10PetMar1
    mkdir /hive/data/genomes/mm10/bed/lastzPetMar1.2012-04-02
    cd /hive/data/genomes/mm10/bed/lastzPetMar1.2012-04-02

    cat << '_EOF_' > DEF
# Mouse vs. lamprey
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: lamprey petMar1
SEQ2_DIR=/scratch/data/petMar1/petMar1.2bit
SEQ2_LEN=/scratch/data/petMar1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=200

BASE=/hive/data/genomes/mm10/bed/lastzPetMar1.2012-04-02
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #	real    77m3.923s
    # broken swarm, continuing:
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -qRepeats=windowmaskerSdust -continue=cat `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > cat.log 2>&1 &
    # missing qRepeats specification
    rm axtChain/mm10.petMar1.net
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        -qRepeats=windowmaskerSdust -continue=load `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > load.log 2>&1 &
    #	real    6m31.527s
    cat fb.mm10.chainPetMar1Link.txt
    #	29205053 bases of 2652783500 (1.101%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzPetMar1.2012-04-02 lastz.petMar1

    #	and for the swap
    mkdir /hive/data/genomes/petMar1/bed/blastz.mm10.swap
    cd /hive/data/genomes/petMar1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzPetMar1.2012-04-02/DEF \
        -qRepeats=windowmaskerSdust -workhorse=hgwdev \
	-smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #   real    17m40.196s
    cat  fb.petMar1.chainMm10Link.txt
    #   26274715 bases of 831696438 (3.159%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/petMar1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
## 60-Way Multiz (DONE - 2011-09-28 - Hiram)
    ssh hgwdev
    mkdir /hive/data/genomes/mm10/bed/multiz60way
    cd /hive/data/genomes/mm10/bed/multiz60way

    # from the 62-way in the source tree, do not need aliMis1 and croPor1:
    /cluster/bin/phast/tree_doctor --prune ailMis1,croPor1 \
        /cluster/home/hiram/kent/src/hg/utils/phyloTrees/62way.nh > 60way.nh

    # note, newer assemblies: susScr3, dasNov3, felCat5, hetGla2, turTru2,
    #   nomLeu2, oreNil2

    #	what that looks like:
    cat 60way.nh
# (((((((((((((((((((hg19:0.006550,panTro4:0.006840):0.002220,
# gorGor3:0.008964):0.009693,ponAbe2:0.018940):0.003471,
# nomLeu2:0.022270):0.012040,(rheMac3:0.007991,
# papHam1:0.008042):0.029610):0.021830,(calJac3:0.030000,
# saiBol1:0.040000):0.039650):0.052090,tarSyr1:0.111400):0.020520,
# (micMur1:0.085600,otoGar3:0.119400):0.020520):0.015494,
# tupBel1:0.186203):0.004937,(((((mm10:0.084509,rn5:0.091589):0.197773,
# dipOrd1:0.211609):0.022992,(hetGla2:0.100000,
# cavPor3:0.125629):0.100000):0.010150,speTri2:0.148468):0.025746,
# (oryCun2:0.114227,ochPri2:0.201069):0.101463):0.015313):0.020593,
# (((susScr3:0.120000,(vicPac1:0.087275,(turTru2:0.064688,
# (oviAri1:0.100000,bosTau7:0.100000):0.023592):0.025153):0.020335):0.020000,
# ((equCab2:0.109397,(felCat5:0.098612,
# (canFam3:0.052458,ailMel1:0.050000):0.050000):0.049845):0.006219,
# (myoLuc2:0.142540,pteVam1:0.113399):0.033706):0.004508):0.011671,
# (eriEur1:0.221785,sorAra1:0.269562):0.056393):0.021227):0.023664,
# ((((loxAfr3:0.082242,proCap1:0.155358):0.026990,echTel1:0.245936):0.010000,
# triMan1:0.100000):0.049697,(dasNov3:0.116664,
# choHof1:0.096357):0.053145):0.006717):0.234728,(monDom5:0.125686,
# (sarHar1:0.100000,macEug2:0.072008):0.050000):0.215100):0.071664,
# ornAna1:0.456592):0.109504,(((((melGal1:0.100000,galGal4:0.065536):0.100000,
# taeGut1:0.171542):0.199223,melUnd1:0.100000):0.155143,
# anoCar2:0.539241):0.122371,chrPic1:0.200000):0.010000):0.050000,
# xenTro3:0.855573):0.100000,latCha1:0.855573):0.311354,
# ((((((tetNig2:0.224159,fr3:0.203847):0.097590,oreNil2:0.200000):0.097590,
# gasAcu1:0.316413):0.030000,oryLat2:0.511970):0.030000,
# gadMor1:0.350000):0.225640,danRer7:0.730752):0.147949):0.526688,
# petMar1:0.526688);

    #	rearrange to get mm10 on top:
    cat << '_EOF_' > mm10.60way.nh
(((((((((((((((mm10:0.084509,rn5:0.091589):0.197773,dipOrd1:0.211609):0.022992,
(hetGla2:0.1,cavPor3:0.125629):0.1):0.01015,speTri2:0.148468):0.025746,(oryCun2:0.114227,ochPri2:0.201069):0.101463):0.015313,
(((((((((hg19:0.00655,panTro4:0.00684):0.00222,gorGor3:0.008964):0.009693,ponAbe2:0.01894):0.003471,
nomLeu2:0.02227):0.01204,(rheMac3:0.007991,papHam1:0.008042):0.02961):0.02183,
(calJac3:0.03,saiBol1:0.04):0.03965):0.05209,tarSyr1:0.1114):0.02052,(micMur1:0.0856,otoGar3:0.1194):0.02052):0.015494,
tupBel1:0.186203):0.004937):0.020593,
((susScr3:0.12,(vicPac1:0.087275,(turTru2:0.064688,
(oviAri1:0.1,bosTau7:0.1):0.023592):0.025153):0.020335):0.01,
((((felCat5:0.098612,
(canFam3:0.052458,ailMel1:0.05):0.05):0.049845,equCab2:0.109397):0.006219,
(myoLuc2:0.14254,pteVam1:0.113399):0.033706):0.004508,(eriEur1:0.221785,
sorAra1:0.269562):0.056393):0.021227):0.01):0.013664,((((loxAfr3:0.082242,proCap1:0.155358):0.02699,
echTel1:0.245936):0.01,triMan1:0.1):0.049697,(dasNov3:0.116664,
choHof1:0.096357):0.053145):0.006717):0.234728,(monDom5:0.125686,(sarHar1:0.1,
macEug2:0.072008):0.05):0.2151):0.071664,ornAna1:0.456592):0.109504,
(((((melGal1:0.1,galGal4:0.065536):0.1,taeGut1:0.171542):0.199223,melUnd1:0.1):0.155143,anoCar2:0.539241):0.122371,
chrPic1:0.2):0.01):0.05,xenTro3:0.855573):0.1,latCha1:0.855573):0.311354,
((((((tetNig2:0.224159,fr3:0.203847):0.09759,oreNil2:0.2):0.09759,gasAcu1:0.316413):0.03,
oryLat2:0.51197):0.03,gadMor1:0.35):0.22564,danRer7:0.730752):0.147949):0.526688,petMar1:0.526688);
'_EOF_'
    # << happy emacs

    # extract species list from that .nh file
    sed 's/[a-z][a-z]*_//g; s/:[0-9\.][0-9\.]*//g; s/;//; /^ *$/d' \
        mm10.60way.nh | xargs echo | sed 's/ //g; s/,/ /g' \
        | sed 's/[()]//g; s/,/ /g' | tr '[ ]' '[\n]' > species.list.txt

    # construct db to name translation list:
    cat species.list.txt | while read DB
do
hgsql -N -e "select name,organism from dbDb where name=\"${DB}\";" hgcentraltest
done | sed -e "s/\t/->/; s/ /_/g;" | sed -e 's/$/;/' | sed -e 's/\./_/g' \
        > db.to.name.txt

    # construct a common name .nh file:
    /cluster/bin/phast/tree_doctor --rename \
    "`cat db.to.name.txt`" mm10.60way.nh | sed -e 's/00*)/)/g; s/00*,/,/g' \
        | sed -e 's/X__trop/X._trop/' > mm10.60way.commonNames.nh
# (((((((((((((((Mouse:0.084509,Rat:0.091589):0.197773,
# Kangaroo_rat:0.211609):0.022992,(Naked_mole:0.1,
# Guinea_pig:0.125629):0.1):0.01015,Squirrel:0.148468):0.025746,
# (Rabbit:0.114227,Pika:0.201069):0.101463):0.015313,
# (((((((((Human:0.00655,Chimp:0.00684):0.00222,Gorilla:0.008964):0.009693,
# Orangutan:0.01894):0.003471,Gibbon:0.02227):0.01204,
# (Chinese_rhesus:0.007991,Baboon:0.008042):0.02961):0.02183,
# (Marmoset:0.03,Squirrel_monkey:0.04):0.03965):0.05209,
# Tarsier:0.1114):0.02052,(Mouse_lemur:0.0856,
# Bushbaby:0.1194):0.02052):0.015494,Tree_shrew:0.186203):0.004937):0.020593,
# ((Pig:0.12,(Alpaca:0.087275,(Dolphin:0.064688,
# (Sheep:0.1,Cow:0.1):0.023592):0.025153):0.020335):0.01,
# ((((Cat:0.098612,(Dog:0.052458,Panda:0.05):0.05):0.049845,
# Horse:0.109397):0.006219,(Microbat:0.14254,
# Megabat:0.113399):0.033706):0.004508,(Hedgehog:0.221785,
# Shrew:0.269562):0.056393):0.021227):0.01):0.013664,
# ((((Elephant:0.082242,Rock_hyrax:0.155358):0.02699,
# Tenrec:0.245936):0.01,Manatee:0.1):0.049697,
# (Armadillo:0.116664,Sloth:0.096357):0.053145):0.006717):0.234728,
# (Opossum:0.125686,(Tasmanian_devil:0.1,
# Wallaby:0.072008):0.05):0.2151):0.071664,Platypus:0.456592):0.109504,
# (((((Turkey:0.1,Chicken:0.065536):0.1,Zebra_finch:0.171542):0.199223,
# Budgerigar:0.1):0.155143,Lizard:0.539241):0.122371,
# Painted_turtle:0.2):0.01):0.05,X._tropicalis:0.855573):0.1,
# Coelacanth:0.855573):0.311354,((((((Tetraodon:0.224159,
# Fugu:0.203847):0.09759,Nile_tilapia:0.2):0.09759,
# Stickleback:0.316413):0.03,Medaka:0.51197):0.03,
# Atlantic_cod:0.35):0.22564,Zebrafish:0.730752):0.147949):0.526688,
# Lamprey:0.526688);


    #	Use this specification in the phyloGif tool:
    #	http://genome.ucsc.edu/cgi-bin/phyloGif
    #	to obtain a png image for src/hg/htdocs/images/phylo/mm10_60way.png

    /cluster/bin/phast/all_dists mm10.60way.nh | grep mm10 \
        | sed -e "s/mm10^I//" | sort -k2n > 60way.distances.txt
    #	Use this output to create the table below
    head 60way.distances.txt
# rn5     0.176098
# speTri2 0.463892
# micMur1 0.483034
# dipOrd1 0.493891
# vicPac1 0.504686
# hetGla2 0.505274
# hg19    0.505328
# gorGor3 0.505522
# panTro4 0.505618
# nomLeu2 0.505664
    cat << '_EOF_' > sizeStats.pl
#!/usr/bin/env perl

use strict;
use warnings;

open (FH, "<60way.distances.txt") or
        die "can not read 60way.distances.txt";

my $count = 0;
while (my $line = <FH>) {
    chomp $line;
    my ($D, $dist) = split('\s+', $line);
    my $chain = "chain" . ucfirst($D);
    my $B="/hive/data/genomes/mm10/bed/lastz.$D/fb.mm10." .
        $chain . "Link.txt";
    my $chainLinkMeasure =
        `awk '{print \$5}' ${B} 2> /dev/null | sed -e "s/(//; s/)//"`;
    chomp $chainLinkMeasure;
    $chainLinkMeasure = 0.0 if (length($chainLinkMeasure) < 1);
    $chainLinkMeasure =~ s/\%//;
    my $swapFile="/hive/data/genomes/${D}/bed/lastz.mm10/fb.${D}.chainMm10Link.txt";
    my $swapMeasure = "N/A";
    if ( -s $swapFile ) {
	$swapMeasure =
	    `awk '{print \$5}' ${swapFile} 2> /dev/null | sed -e "s/(//; s/)//"`;
	chomp $swapMeasure;
	$swapMeasure = 0.0 if (length($swapMeasure) < 1);
	$swapMeasure =~ s/\%//;
    }
    my $orgName=
    `hgsql -N -e "select organism from dbDb where name='$D';" hgcentraltest`;
    chomp $orgName;
    if (length($orgName) < 1) {
        $orgName="N/A";
    }
    ++$count;
    printf "# %02d  %.4f (%% %06.3f) (%% %06.3f) - %s %s\n", $count, $dist,
        $chainLinkMeasure, $swapMeasure, $orgName, $D;
}
close (FH);
'_EOF_'
    # << happy emacs
    chmod +x ./sizeStats.pl
    ./sizeStats.pl
#

#	If you can fill in all the numbers in this table, you are ready for
#	the multiple alignment procedure

#       featureBits chainLink measures
#               chainAnoCar2Link
#  N distance  on mm10  on other     other species
# 01  0.1761 (% 67.353) (% 70.278) - Rat rn5
# 02  0.4639 (% 34.217) (% 39.244) - Squirrel speTri2
# 03  0.4830 (% 26.636) (% 37.574) - Mouse lemur micMur1
# 04  0.4939 (% 19.460) (% 27.512) - Kangaroo rat dipOrd1
# 05  0.5047 (% 22.636) (% 31.769) - Alpaca vicPac1
# 06  0.5053 (% 32.753) (% 37.989) - Naked mole rat hetGla2
# 07  0.5053 (% 38.226) (% 35.249) - Human hg19
# 08  0.5055 (% 33.987) (% 34.349) - Gorilla gorGor3
# 09  0.5056 (% 34.674) (% 31.924) - Chimp panTro4
# 10  0.5057 (% 34.031) (% 32.274) - Gibbon nomLeu2
# 11  0.5058 (% 34.496) (% 30.610) - Orangutan ponAbe2
# 12  0.5073 (% 30.267) (% 33.492) - Dolphin turTru2
# 13  0.5088 (% 24.560) (% 24.986) - Tarsier tarSyr1
# 14  0.5090 (% 33.931) (% 33.464) - Chinese rhesus rheMac3
# 15  0.5090 (% 33.577) (% 32.023) - Baboon papHam1
# 16  0.5168 (% 29.795) (% 32.926) - Bushbaby otoGar3
# 17  0.5171 (% 25.685) (% 29.445) - Pig susScr3
# 18  0.5192 (% 32.450) (% 31.301) - Marmoset calJac3
# 19  0.5284 (% 34.415) (% 37.138) - Horse equCab2
# 20  0.5292 (% 32.339) (% 33.848) - Squirrel monkey saiBol1
# 21  0.5309 (% 28.447) (% 29.115) - Guinea pig cavPor3
# 22  0.5470 (% 18.019) (% 23.687) - Sloth choHof1
# 23  0.5472 (% 26.546) (% 24.649) - Manatee triMan1
# 24  0.5476 (% 19.766) (% 25.144) - Tree shrew tupBel1
# 25  0.5569 (% 25.248) (% 25.677) - Rabbit oryCun2
# 26  0.5599 (% 27.345) (% 38.627) - Megabat pteVam1
# 27  0.5662 (% 26.255) (% 25.383) - Cow bosTau7
# 28  0.5662 (% 15.341) (% 31.925) - Sheep oviAri1
# 29  0.5664 (% 25.823) (% 21.616) - Elephant loxAfr3
# 30  0.5673 (% 25.201) (% 21.066) - Armadillo dasNov3
# 31  0.5675 (% 29.725) (% 32.244) - Cat felCat5
# 32  0.5689 (% 30.979) (% 35.562) - Panda ailMel1
# 33  0.5713 (% 29.144) (% 31.624) - Dog canFam3
# 34  0.5891 (% 24.363) (% 33.650) - Microbat myoLuc2
# 35  0.6395 (% 15.147) (% 16.214) - Rock hyrax proCap1
# 36  0.6437 (% 14.542) (% 19.908) - Pika ochPri2
# 37  0.6865 (% 09.856) (% 12.264) - Hedgehog eriEur1
# 38  0.7031 (% 10.947) (% 14.117) - Tenrec echTel1
# 39  0.7343 (% 09.382) (% 13.569) - Shrew sorAra1
# 40  0.9626 (% 04.353) (% 04.448) - Wallaby macEug2
# 41  0.9663 (% 09.584) (% 07.205) - Opossum monDom5
# 42  0.9906 (% 08.479) (% 07.888) - Tasmanian devil sarHar1
# 43  1.0166 (% 04.731) (% 05.488) - Painted turtle chrPic1
# 44  1.1537 (% 05.348) (% 07.334) - Platypus ornAna1
# 45  1.1942 (% 03.589) (% 07.350) - Budgerigar melUnd1
# 46  1.4589 (% 03.676) (% 08.100) - Chicken galGal4
# 47  1.4649 (% 03.599) (% 07.304) - Zebra finch taeGut1
# 48  1.4782 (% 03.331) (% 04.988) - Lizard anoCar2
# 49  1.4934 (% 03.511) (% 08.211) - Turkey melGal1
# 50  1.7122 (% 03.125) (% 06.651) - X. tropicalis xenTro3
# 51  1.8122 (% 02.715) (% 03.380) - Coelacanth latCha1
# 52  1.9916 (% 01.726) (% 06.810) - Atlantic cod gadMor1
# 53  1.9992 (% 01.957) (% 06.091) - Nile tilapia oreNil2
# 54  2.0180 (% 02.016) (% 10.927) - Stickleback gasAcu1
# 55  2.1006 (% 01.789) (% 12.134) - Fugu fr3
# 56  2.1209 (% 01.735) (% 13.642) - Tetraodon tetNig2
# 57  2.1467 (% 02.602) (% 05.107) - Zebrafish danRer7
# 58  2.1835 (% 01.936) (% 06.561) - Medaka oryLat2
# 59  2.3214 (% 01.101) (% 03.159) - Lamprey petMar1

# None of this concern for distances matters in building the first step, the
# maf files.

    # create species list and stripped down tree for autoMZ
    sed 's/[a-z][a-z]*_//g; s/:[0-9\.][0-9\.]*//g; s/;//; /^ *$/d' \
	mm10.60way.nh > tmp.nh
    echo `cat tmp.nh` > tree-commas.nh
    echo `cat tree-commas.nh` | sed 's/ //g; s/,/ /g' > tree.nh
    sed 's/[()]//g; s/,/ /g' tree.nh > species.list

    #	bash shell syntax here ...
    cd /hive/data/genomes/mm10/bed/multiz60way
    export H=/hive/data/genomes/mm10/bed
    mkdir mafLinks
    for G in `sed -e "s/mm10 //" species.list`
    do
	mkdir mafLinks/$G
	if [ -s ${H}/lastz.${G}/mafRBestNet/chr1.maf.gz ]; then
	    echo "$G - recipBest"
	    ln -s ${H}/lastz.$G/mafRBestNet/*.maf.gz ./mafLinks/$G
	else
	    if [ -s ${H}/lastz.${G}/mafSynNet/chr1.maf.gz ]; then
		echo "$G - synNet"
		ln -s ${H}/lastz.$G/mafSynNet/*.maf.gz ./mafLinks/$G
	    else
		if [ -s ${H}/lastz.${G}/mafNet/chr1.maf.gz ]; then
		    echo "$G - mafNet"
		    ln -s ${H}/lastz.$G/mafNet/*.maf.gz ./mafLinks/$G
		else
		    echo "missing directory lastz.${G}/*Net"
		fi
	    fi
	fi
    done

    #	verify the alignment type is correct:
    for D in `grep -v mm10 /hive/users/hiram/bigWays/mm10.60way/ordered.list`
do
    ls -l mafLinks/$D/chr1.maf.gz | awk '{print $NF}'
done
    #	compare to the list at:
    #	http://genomewiki.ucsc.edu/index.php/Mm10_Genome_size_statistics

    #	need to split these things up into smaller pieces for
    #	efficient kluster run.
    cd /hive/data/genomes/mm10/bed/multiz60way
    mkdir mafSplit
    cd mafSplit
    #	mafSplitPos splits on gaps or repeat areas that will not have
    #	any chains, approx 5 Mbp intervals, gaps at least 10,000
    mafSplitPos -minGap=10000 mm10 5 stdout | sort -u \
	| sort -k1,1 -k2,2n > mafSplit.bed
    #	There is a splitRegions.pl script here (copied from previous hg19 46way)
    #	that can create a custom track from this mafSplit.bed file.
    #	Take a look at that in the browser and see if it looks OK,
    #	check the number of sections on each chrom to verify none are
    #	too large.  Despite the claim above, it does appear that some
    #	areas are split where actual chains exist.
    ./splitRegions.pl mafSplit.bed > splitRegions.ct

    # to see the sizes of the regions:
    grep "^chr" splitRegions.ct | awk '{print $3-$2,$0}' | sort -rn | less

    #	run a kluster job to split them all
    ssh swarm
    cd /hive/data/genomes/mm10/bed/multiz60way/mafSplit
    cat << '_EOF_' > runOne
#!/bin/csh -ef
set G = $1
set C = $2
mkdir -p $G
pushd $G > /dev/null
if ( -s ../../mafLinks/${G}/${C}.maf.gz ) then
    if ( -s mm10_${C}.00.maf ) then
        /bin/rm -f mm10_${C}.*.maf
    endif
    /cluster/bin/x86_64/mafSplit ../mafSplit.bed mm10_ ../../mafLinks/${G}/${C}.maf.gz
    /bin/gzip mm10_${C}.*.maf
else
    /bin/touch mm10_${C}.00.maf
    /bin/gzip mm10_${C}.00.maf
endif
popd > /dev/null
'_EOF_'
    # << happy emacs
    chmod +x runOne

    cat << '_EOF_' > template
#LOOP
runOne $(root1) $(root2) {check out exists+ $(root1)/mm10_$(root2).00.maf.gz}
#ENDLOOP
'_EOF_'
    # << happy emacs

    for G in `sed -e "s/mm10 //" ../species.list`
do
    echo $G
done > species.list
    cut -f 1 ../../../chrom.sizes > chr.list

    gensub2 species.list chr.list template jobList
    para -ram=8g create jobList
    para try ... check ... push ... etc...
# Completed: 3894 of 3894 jobs
# CPU time in finished jobs:      18929s     315.49m     5.26h    0.22d  0.001 y
# IO & Wait Time:                 62908s    1048.46m    17.47h    0.73d  0.002 y
# Average job time:                  21s       0.35m     0.01h    0.00d
# Longest finished job:             346s       5.77m     0.10h    0.00d
# Submission to last job:           471s       7.85m     0.13h    0.01d

    # construct a list of all possible maf file names.
    # they do not all exist in each of the species directories
    find . -type f | grep "maf.gz" | wc -l
    # 19733
    find . -type f | grep ".maf.gz$" | xargs -L 1 basename | sort -u > maf.list
    wc -l maf.list
    #   336 maf.list

    mkdir /hive/data/genomes/mm10/bed/multiz60way/splitRun
    cd /hive/data/genomes/mm10/bed/multiz60way/splitRun
    mkdir maf run
    cd run
    mkdir penn
    cp -p /cluster/bin/penn/multiz.2009-01-21/multiz penn
    cp -p /cluster/bin/penn/multiz.2009-01-21/maf_project penn
    cp -p /cluster/bin/penn/multiz.2009-01-21/autoMZ penn

    #	set the db and pairs directories here
    cat > autoMultiz.csh << '_EOF_'
#!/bin/csh -ef
set db = mm10
set c = $1
set result = $2
set run = `/bin/pwd`
set tmp = /scratch/tmp/$db/multiz.$c
set pairs = /hive/data/genomes/mm10/bed/multiz60way/mafSplit
/bin/rm -fr $tmp
/bin/mkdir -p $tmp
/bin/cp -p ../../tree.nh ../../species.list $tmp
pushd $tmp > /dev/null
foreach s (`/bin/sed -e "s/$db //" species.list`)
    set in = $pairs/$s/$c
    set out = $db.$s.sing.maf
    if (-e $in.gz) then
        /bin/zcat $in.gz > $out
        if (! -s $out) then
            echo "##maf version=1 scoring=autoMZ" > $out
        endif
    else if (-e $in) then
        /bin/ln -s $in $out
    else
        echo "##maf version=1 scoring=autoMZ" > $out
    endif
end
set path = ($run/penn $path); rehash
$run/penn/autoMZ + T=$tmp E=$db "`cat tree.nh`" $db.*.sing.maf $c \
        > /dev/null
popd > /dev/null
/bin/rm -f $result
/bin/cp -p $tmp/$c $result
/bin/rm -fr $tmp
'_EOF_'
# << happy emacs
    chmod +x autoMultiz.csh

    cat  << '_EOF_' > template
#LOOP
./autoMultiz.csh $(root1) {check out line+ /hive/data/genomes/mm10/bed/multiz60way/splitRun/maf/$(root1)}
#ENDLOOP
'_EOF_'
# << happy emacs

    ln -s ../../mafSplit/maf.list maf.list
    ssh swarm
    cd /hive/data/genomes/mm10/bed/multiz60way/splitRun/run
    # the tac reverses the list to get the small jobs first
    gensub2 maf.list single template stdout | tac > jobList
    para -ram=8g create jobList
# Completed: 336 of 336 jobs
# CPU time in finished jobs:    2828651s   47144.19m   785.74h   32.74d  0.090 y
# IO & Wait Time:                200533s    3342.21m    55.70h    2.32d  0.006 y
# Average job time:                9015s     150.26m     2.50h    0.10d
# Longest finished job:           47029s     783.82m    13.06h    0.54d
# Submission to last job:         48982s     816.37m    13.61h    0.57d

    # put the split maf results back together into a single maf file
    #	eliminate duplicate comments
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/splitRun
    mkdir ../maf
    #	the sed edits take out partitioning name information from the comments
    #	so the multiple parts will condense to smaller number of lines
    #	this takes almost 2 hours of time, resulting in a bit over 150 Gb,
    #	almost all chrom files over 1 Gb, up to almost 10 Gb for chr2
    #	HOWEVER, this is actually not necessary to maintain these comments,
    #	they are lost during the mafAddIRows

    cat << '_EOF_' >> runOne
#!/bin/csh -fe
set C = $1
if ( -s ../maf/${C}.maf.gz ) then
    rm -f ../maf/${C}.maf.gz
endif
head -q -n 1 maf/mm10_${C}.*.maf | sort -u > ../maf/${C}.maf
grep -h "^#" maf/mm10_${C}.*.maf | egrep -v "maf version=1|eof maf" | \
    sed -e "s#${C}.[0-9][0-9]*#${C}#g; s#_MZ_[^ ]* # #g;" \
        | sort -u >> ../maf/${C}.maf
grep -h -v "^#" `ls maf/mm10_${C}.*.maf | sort -t. -k2,2n` >> ../maf/${C}.maf
tail -q -n 1 maf/mm10_${C}.*.maf | sort -u >> ../maf/${C}.maf
'_EOF_'
    # << happy emacs
    chmod +x runOne

    cat << '_EOF_' >> template
#LOOP
runOne $(root1) {check out exists+ ../maf/$(root1).maf}
#ENDLOOP
'_EOF_'
    # << happy emacs

    cut -f1 ../../../chrom.sizes > chr.list
    ssh encodek
    cd /hive/data/genomes/mm10/bed/multiz60way/splitRun
    gensub2 chr.list single template jobList
    para -ram=8g create jobList
    para try ... check ... push ... etc ...
# Completed: 62 of 66 jobs
# Crashed: 4 jobs
# CPU time in finished jobs:        461s       7.68m     0.13h    0.01d  0.000 y
# IO & Wait Time:                 17863s     297.72m     4.96h    0.21d  0.001 y
# Average job time:                 296s       4.93m     0.08h    0.00d
# Longest finished job:            1144s      19.07m     0.32h    0.01d
# Submission to last job:          1156s      19.27m     0.32h    0.01d

    # these four have empty results:
#       chrUn_GL456383
#       chrUn_GL456389
#       chrUn_GL456390
#       chrUn_GL456396

    # Load into database
    ssh hgwdev
    mkdir -p /gbdb/mm10/multiz60way
    cd /hive/data/genomes/mm10/bed/multiz60way/maf
    ln -s `pwd`/*.maf /gbdb/mm10/multiz60way

    # this generates an immense multiz60way.tab file in the directory
    #	where it is running.  Best to run this over in scratch.
    #   This is going to take all day.
    cd /scratch/tmp
    time nice -n +19 hgLoadMaf mm10 multiz60way
    #   Loaded 56185270 mafs in 66 files from /gbdb/mm10/multiz60way
    #   real    72m45.513s
# -rw-rw-r-- 1 2857704841 Apr 18 10:49 multiz60way.tab

    time cat /gbdb/mm10/multiz60way/*.maf \
        | nice -n +19 hgLoadMafSummary -verbose=2 -minSize=30000 \
	-mergeGap=1500 -maxSize=200000 mm10 multiz60waySummary stdin
    #   Created 12012784 summary blocks from 1074134156 components and
    #   56185270 mafs from stdin
    #   real    104m2.107s

    wc -l multiz60way*.tab
    #   56185270 multiz60way.tab
    #   12012784 multiz60waySummary.tab
    #   68198054 total
    #   -rw-rw-r-- 1 2857704841 Apr 18 10:49 multiz60way.tab
    #   -rw-rw-r-- 1  567210414 Apr 18 17:28 multiz60waySummary.tab

    rm multiz60way*.tab

#######################################################################
# GAP ANNOTATE MULTIZ9WAY MAF AND LOAD TABLES (DONE - 2012-05-31 - Hiram)
    # mafAddIRows has to be run on single chromosome maf files, it does not
    #	function correctly when more than one reference sequence
    #	are in a single file.
    mkdir -p /hive/data/genomes/mm10/bed/multiz60way/anno
    cd /hive/data/genomes/mm10/bed/multiz60way/anno

    cd /hive/data/genomes/mm10/bed/multiz60way/anno
    # check for N.bed files everywhere:
    for DB in `cat ../species.list`
do
    if [ ! -s /hive/data/genomes/${DB}/${DB}.N.bed ]; then
        echo "MISS: ${DB}"
        cd /hive/data/genomes/${DB}
        twoBitInfo -nBed ${DB}.2bit ${DB}.N.bed
    else
        echo "  OK: ${DB}"
    fi
done

    cd /hive/data/genomes/mm10/bed/multiz60way/anno
    for DB in `cat ../species.list`
do
    echo "${DB} "
    ln -s  /hive/data/genomes/${DB}/${DB}.N.bed ${DB}.bed
    echo ${DB}.bed  >> nBeds
    ln -s  /hive/data/genomes/${DB}/chrom.sizes ${DB}.len
    echo ${DB}.len  >> sizes
done
    # make sure they all are successful symLinks:
    ls -ogrtL

    screen -S mm10      # use a screen to control this longish job
    ssh swarm
    cd /hive/data/genomes/mm10/bed/multiz60way/anno
    mkdir result
    # NEXT TIME: this template should have a check out exists+ statement
    cat << '_EOF_' > template
#LOOP
mafAddIRows -nBeds=nBeds $(path1) /hive/data/genomes/mm10/mm10.2bit {check out line+ result/$(file1)}
#ENDLOOP
'_EOF_'
    # << happy emacs

    ls ../maf/*.maf > maf.list
    # the tac puts the short jobs first
    gensub2 maf.list single template stdout | tac > jobList
    # limit jobs to one per node with the ram=8g requirement
    para -ram=8g create jobList
    para try ... check ... push ...
# Completed: 46 of 66 jobs
# CPU time in finished jobs:        350s       5.83m     0.10h    0.00d  0.000 y
# IO & Wait Time:                   603s      10.06m     0.17h    0.01d  0.000 y
# Average job time:                  21s       0.35m     0.01h    0.00d
# Longest finished job:              54s       0.90m     0.01h    0.00d
# Submission to last job:           113s       1.88m     0.03h    0.00d

    # a number of these jobs did not finish due to memory limitations.
    # The jobs would sit on the nodes appearing to occupy 8 Gb of memory,
    # but did not see any swapping or CPU time accumulation.  Stop the
    # batch and run the rest manually on hgwdev:
#!/bin/sh

export maxMem=188743680
ulimit -S -m $maxMem -v $maxMem

mafAddIRows -nBeds=nBeds ../maf/chrX.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chrX.maf &
mafAddIRows -nBeds=nBeds ../maf/chr9.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chr9.maf &
mafAddIRows -nBeds=nBeds ../maf/chr8.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chr8.maf &
mafAddIRows -nBeds=nBeds ../maf/chr7.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chr7.maf &
wait
mafAddIRows -nBeds=nBeds ../maf/chr6.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chr6.maf &
mafAddIRows -nBeds=nBeds ../maf/chr5.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chr5.maf &
mafAddIRows -nBeds=nBeds ../maf/chr4.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chr4.maf &
mafAddIRows -nBeds=nBeds ../maf/chr3.maf /hive/data/genomes/mm10/mm10.2bit hgwdev/chr3.maf &
wait
... etc ...
    # the run time for those 20 jobs:
    #   real    159m49.217s

    # verify all result files have some content, look for 0 size files:
    find . -type f -size 0
    # should see none

    # combine into one file  (realized after this, that we do *not* need
    #                           this single file.  Individual files are OK.
    head -q -n 1 result/chrM.maf > mm10.60way.maf
    time for F in hgwdev/*.maf result/*.maf
do
    grep -h -v "^#" ${F}
done >> mm10.60way.maf
    #   real    1082m47.484s -> 18 hours !
# -rw-rw-r-- 1 261567878241 Jun  8 10:30 mm10.60way.maf
    du -hsc mm10.60way.maf
    #   244G    mm10.60way.maf

    #	these maf files do not have the end marker, this does nothing:
    #	tail -q -n 1 result/chrM.maf >> mm10.60way.maf
    # How about an official end marker:
    echo "##eof maf" >> mm10.60way.maf

    # construct symlinks to get the individual maf files into gbdb:
    mkdir /gbdb/mm10/multiz60way/maf
    ln -s `pwd`/result/*.maf `pwd`/hgwdev/*.maf /gbdb/mm10/multiz60way/maf/

    # Load into database
    rm /gbdb/mm10/multiz60way/*.maf   # remove previous results
    cd /scratch/tmp
    time nice -n +19 hgLoadMaf -pathPrefix=/gbdb/mm10/multiz60way/maf \
        mm10 multiz60way
    #   Loaded 58087742 mafs in 66 files from /gbdb/mm10/multiz60way/maf
    #   real    868m28.108s

    time (cat /gbdb/mm10/multiz60way/maf/*.maf \
        | hgLoadMafSummary -verbose=2 -minSize=30000 \
	-mergeGap=1500 -maxSize=200000 mm10 multiz60waySummary stdin)

    #   -rw-rw-r-- 1 3009209972 Jun  9 03:23 multiz60way.tab
    #   -rw-rw-r-- 1  591235982 Jun 11 18:34 multiz60waySummary.tab

    rm multiz60way*.tab

#######################################################################
# MULTIZ60WAY MAF FRAMES (DONE - 2012-05-30 - Hiram)
    ssh hgwdev
    mkdir /hive/data/genomes/mm10/bed/multiz60way/frames
    cd /hive/data/genomes/mm10/bed/multiz60way/frames
#   survey all the genomes to find out what kinds of gene tracks they have
    cat << '_EOF_' > showGenes.csh
#!/bin/csh -fe
foreach db (`cat ../species.list`)
    echo -n "${db}: "
    set tables = `hgsql $db -N -e "show tables like '%Gene%'"`
    foreach table ($tables)
        if ($table == "ensGene" || $table == "refGene" || \
           $table == "mgcGenes" || $table == "knownGene" || \
           $table == "xenoRefGene" ) then
           set count = `hgsql $db -N -e "select count(*) from $table"`
            echo -n "${table}: ${count}, "
        endif
    end
    set orgName = `hgsql hgcentraltest -N -e \
            "select scientificName from dbDb where name='$db'"`
    set orgId = `hgsql hg19 -N -e \
            "select id from organism where name='$orgName'"`
    if ($orgId == "") then
        echo "Mrnas: 0"
    else
        set count = `hgsql hg19 -N -e "select count(*) from gbCdnaInfo where organism=$orgId"`
        echo "Mrnas: ${count}"
    endif
end
'_EOF_'
    # << happy emacs
    chmod +x ./showGenes.csh
    time ./showGenes.csh > showGenes.txt
    #   real    9m11.678s

    #   rearrange that output to create four sections, and place these names
    #           in .list files here:
    #   1. knownGene: hg19
    #   2. refGene: bosTau7 danRer7 galGal4 mm10 rheMac3 rn5 susScr3 xenTro3
    #   3. ensGene: ailMel1 anoCar2 calJac3 cavPor3 choHof1 dipOrd1 echTel1
    #               equCab2 eriEur1 fr3 gasAcu1 gorGor3 loxAfr3 melGal1
    #               micMur1 monDom5 myoLuc2 ochPri2 ornAna1 oryCun2 oryLat2
    #               panTro4 ponAbe2 proCap1 pteVam1 sorAra1 taeGut1 tarSyr1
    #               tetNig2 tupBel1 vicPac1
    #   4. xenoRefGene: canFam3 chrPic1 dasNov3 felCat5 hetGla2 latCha1 macEug2
    #               nomLeu2 otoGar3 oviAri1 papHam1 petMar1 saiBol1 sarHar1
    #               triMan1
    #   5. genscan: gadMor1 melUnd1 oreNil2 speTri2 turTru2

    mkdir genes
    #   1. knownGene: hg19
    hgsql -N -e "select name,chrom,strand,txStart,txEnd,cdsStart,cdsEnd,exonCount,exonStarts,exonEnds from knownGene" hg19 \
      | genePredSingleCover stdin stdout | gzip -2c \
        > genes/hg19.gp.gz
    #   2. refGene, want the full extended genePred:
    for DB in `cat refGene.list`
do
hgsql -N -e "select * from refGene" ${DB} | cut -f2- \
      | genePredSingleCover stdin stdout | gzip -2c \
        > /scratch/tmp/${DB}.tmp.gz
    mv /scratch/tmp/${DB}.tmp.gz genes/$DB.gp.gz
    echo "${DB} done"
done
    #   3. ensGene, want the full extended genePred:
    for DB in `cat ensGene.list`
do
hgsql -N -e "select * from ensGene" ${DB} | cut -f2- \
      | genePredSingleCover stdin stdout | gzip -2c \
        > /scratch/tmp/${DB}.tmp.gz
    mv /scratch/tmp/${DB}.tmp.gz genes/$DB.gp.gz
    echo "${DB} done"
done
    #   4. xenoRefGene, want the full extended genePred:
    for DB in `cat xenoRG.list`
do
hgsql -N -e "select * from xenoRefGene" ${DB} | cut -f2- \
      | genePredSingleCover stdin stdout | gzip -2c \
        > /scratch/tmp/${DB}.tmp.gz
    mv /scratch/tmp/${DB}.tmp.gz genes/$DB.gp.gz
    echo "${DB} done"
done
    #   5. genscan: gadMor1 melUnd1 oreNil2 speTri2 turTru2
    for DB in `cat genscan.list`
do
hgsql -N -e "select * from xenoRefGene" ${DB} | cut -f2- \
      | genePredSingleCover stdin stdout | gzip -2c \
        > /scratch/tmp/${DB}.tmp.gz
    mv /scratch/tmp/${DB}.tmp.gz genes/$DB.gp.gz
    echo "${DB} done"
done

    # verify counts for genes are reasonable:
    for T in genes/*.gz
do
    echo -n "# $T: "
    zcat $T | cut -f1 | sort | uniq -c | wc -l
done

# genes/ailMel1.gp.gz: 19204
# genes/anoCar2.gp.gz: 17766
# genes/bosTau7.gp.gz: 12958
# genes/calJac3.gp.gz: 20843
# genes/canFam3.gp.gz: 20652
# genes/cavPor3.gp.gz: 18631
# genes/choHof1.gp.gz: 12403
# genes/chrPic1.gp.gz: 19433
# genes/danRer7.gp.gz: 13902
# genes/dasNov3.gp.gz: 29551
# genes/dipOrd1.gp.gz: 15784
# genes/echTel1.gp.gz: 16499
# genes/equCab2.gp.gz: 20403
# genes/eriEur1.gp.gz: 11712
# genes/felCat5.gp.gz: 19512
# genes/fr3.gp.gz: 18014
# genes/gadMor1.gp.gz: 27572
# genes/galGal4.gp.gz: 4892
# genes/gasAcu1.gp.gz: 20631
# genes/gorGor3.gp.gz: 20759
# genes/hetGla2.gp.gz: 25749
# genes/hg19.gp.gz: 20718
# genes/latCha1.gp.gz: 18786
# genes/loxAfr3.gp.gz: 19986
# genes/macEug2.gp.gz: 26006
# genes/melGal1.gp.gz: 14050
# genes/melUnd1.gp.gz: 15296
# genes/micMur1.gp.gz: 16240
# genes/mm10.gp.gz: 20985
# genes/monDom5.gp.gz: 19188
# genes/myoLuc2.gp.gz: 19685
# genes/nomLeu2.gp.gz: 22996
# genes/ochPri2.gp.gz: 15970
# genes/oreNil2.gp.gz: 18636
# genes/ornAna1.gp.gz: 17728
# genes/oryCun2.gp.gz: 18921
# genes/oryLat2.gp.gz: 19576
# genes/otoGar3.gp.gz: 24061
# genes/oviAri1.gp.gz: 17890
# genes/panTro4.gp.gz: 18647
# genes/papHam1.gp.gz: 27842
# genes/petMar1.gp.gz: 11089
# genes/ponAbe2.gp.gz: 19895
# genes/proCap1.gp.gz: 16043
# genes/pteVam1.gp.gz: 16966
# genes/rheMac3.gp.gz: 5580
# genes/rn5.gp.gz: 16393
# genes/saiBol1.gp.gz: 23419
# genes/sarHar1.gp.gz: 20694
# genes/sorAra1.gp.gz: 13156
# genes/speTri2.gp.gz: 22377
# genes/susScr3.gp.gz: 3771
# genes/taeGut1.gp.gz: 17354
# genes/tarSyr1.gp.gz: 13615
# genes/tetNig2.gp.gz: 19539
# genes/triMan1.gp.gz: 19514
# genes/tupBel1.gp.gz: 15407
# genes/turTru2.gp.gz: 28375
# genes/vicPac1.gp.gz: 11754
# genes/xenTro3.gp.gz: 8447

    # kluster job to annotate each maf file
    screen -S mm10      # manage long running procedure with screen
    ssh swarm
    cd /hive/data/genomes/mm10/bed/multiz60way/frames
    cat << '_EOF_' > runOne
#!/bin/csh -fe

set C = $1
set G = $2

cat ../maf/${C}.maf | genePredToMafFrames mm10 stdin stdout \
        ${G} genes/${G}.gp.gz | gzip > parts/${C}.${G}.mafFrames.gz
'_EOF_'
    # << happy emacs
    chmod +x runOne

    # older instructions excluded mm10 from the gene.list
    #   this was a mistake.  mm10 can be annotated too.
    #   Mistakenly did this the first run through, had to manually
    #   do the mm10 genes separately on hgwdev after this was done
    ls ../maf | sed -e "s/.maf//" > chr.list
    ls genes | sed -e "s/.gp.gz//" > gene.list

    cat << '_EOF_' > template
#LOOP
runOne $(root1) $(root2) {check out exists+ parts/$(root1).$(root2).mafFrames.gz}
#ENDLOOP
'_EOF_'
    # << happy emacs

    mkdir parts
    gensub2 chr.list gene.list template jobList
    para -ram=8g create jobList
    para try ... check ... push
# Completed: 3960 of 3960 jobs
# CPU time in finished jobs:      85610s    1426.83m    23.78h    0.99d  0.003 y
# IO & Wait Time:               2030956s   33849.27m   564.15h   23.51d  0.064 y
# Average job time:                 534s       8.91m     0.15h    0.01d
# Longest finished job:            3877s      64.62m     1.08h    0.04d
# Submission to last job:         12974s     216.23m     3.60h    0.15d

    # collect all results into one file:
    cd /hive/data/genomes/mm10/bed/multiz60way/frames
    find ./parts -type f | while read F
do
    zcat ${F}
done | sort -k1,1 -k2,2n > multiz60wayFrames.bed
    #   -rw-rw-r-- 1 1164299719 May 30 11:28 multiz60wayFrames.bed

    # verify there are frames on everything:
    cut -f4 multiz60wayFrames.bed | sort | uniq -c | sort -n \
        > annotation.survey.txt
    # should be 60 species:
    wc -l annotation.survey.txt
    #   60 annotation.survey.txt
    # and the minimum numbers:
    head annotation.survey.txt
    #   43900 susScr3
    #   59839 rheMac3
    #   153246 petMar1
    #   162501 choHof1
    # ... etc ...

    #   load the resulting file
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/frames
    time gzip multiz60wayFrames.bed
    #   real    0m51.826s
    # reloading this table 2012-10-11 with more accurate frames:
    time hgLoadMafFrames mm10 multiz60wayFrames multiz60wayFrames.bed.gz
    #   real    3m2.449s
    time featureBits -countGaps mm10 multiz60wayFrames
    #   57707702 bases of 2730871774 (2.113%) in intersection
    #   real    1m45.141s

    #   enable the trackDb entries:
# frames multiz60wayFrames
# irows on
    #   appears to work OK

#########################################################################
# Phylogenetic tree from 60-way (DONE - 2012-05-31 - 2012-06-12 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/multiz60way/4d
    cd /hive/data/genomes/mm10/bed/multiz60way/4d

    # the annotated maf's are in:
    ../anno/result/*.maf

    # using ensGene for mm10, only transcribed genes and nothing
    #	from the randoms and other misc.
    hgsql mm10 -Ne \
    "select * from ensGene WHERE cdsEnd > cdsStart;" | cut -f 2-20 \
	| egrep -E -v "chrM|chrUn|random|_hap" > ensGene.gp
    wc -l *.gp
    #   55423 ensGene.gp

    genePredSingleCover ensGene.gp stdout | sort > ensGeneNR.gp
    wc -l ensGeneNR.gp
    #	22457 ensGeneNR.gp

    ssh encodek
    mkdir /hive/data/genomes/mm10/bed/multiz60way/4d/run
    cd /hive/data/genomes/mm10/bed/multiz60way/4d/run
    mkdir ../mfa

    # newer versions of msa_view have a slightly different operation
    # the sed of the gp file inserts the reference species in the chr name
    cat << '_EOF_' > 4d.csh
#!/bin/csh -fe
set PHASTBIN = /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin
set r = "/hive/data/genomes/mm10/bed/multiz60way"
set c = $1
set infile = $r/anno/result/$2
set outfile = $3
cd /scratch/tmp
# 'clean' maf
perl -wpe 's/^s ([^.]+)\.\S+/s $1/' $infile > $c.maf
awk -v C=$c '$2 == C {print}' $r/4d/ensGeneNR.gp | sed -e "s/\t$c\t/\tmm10.$c\t/" > $c.gp
set NL=`wc -l $c.gp| gawk '{print $1}'`
if ("$NL" != "0") then
    $PHASTBIN/msa_view --4d --features $c.gp -i MAF $c.maf -o SS > $c.ss
    $PHASTBIN/msa_view -i SS --tuple-size 1 $c.ss > $r/4d/run/$outfile
else
    echo "" > $r/4d/run/$outfile
endif
rm -f $c.gp $c.maf $c.ss
'_EOF_'
    # << happy emacs
    chmod +x 4d.csh

    ls -1S /hive/data/genomes/mm10/bed/multiz60way/anno/result/*.maf \
	| sed -e "s#.*multiz60way/anno/result/##" \
	> maf.list

    cat << '_EOF_' > template
#LOOP
4d.csh $(root1) $(path1) {check out line+ ../mfa/$(root1).mfa}
#ENDLOOP
'_EOF_'
    # << happy emacs

    # the tac puts the quick jobs at the front
    gensub2 maf.list single template stdout | tac > jobList
    para create jobList
    para try ... check
    para -maxJob=5 push
    para time
# Completed: 66 of 66 jobs
# CPU time in finished jobs:      13176s     219.60m     3.66h    0.15d  0.000 y
# IO & Wait Time:                 31790s     529.84m     8.83h    0.37d  0.001 y
# Average job time:                 681s      11.36m     0.19h    0.01d
# Longest finished job:            2883s      48.05m     0.80h    0.03d
# Submission to last job:          2925s      48.75m     0.81h    0.03d

    # combine mfa files
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/4d
    # remove the broken empty files, size 0 and size 1:
    find ./mfa -type f -size 0 | xargs rm -f
    # most interesting, this did not identify files of size 1:
#    find ./mfa -type f -size 1
    ls -og mfa | awk '$3 == 1' | awk '{print $NF}' > empty.list
    sed -e "s#^#mfa/##" empty.list | xargs rm -f
    #want comma-less species.list
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/msa_view \
	--aggregate "`cat ../species.list`" mfa/*.mfa | sed s/"> "/">"/ \
	    > 4d.all.mfa
    # check they are all in there:
    grep "^>" 4d.all.mfa | wc -l
    #   60

    # use phyloFit to create tree model (output is phyloFit.mod)
    time nice -n +19 \
	/cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/phyloFit \
	    --EM --precision MED --msa-format FASTA --subst-mod REV \
		--tree ../tree-commas.nh 4d.all.mfa
    #   real    98m59.203s
    mv phyloFit.mod all.mod

    grep TREE all.mod
#TREE: (((((((((((((((mm10:0.0855383,rn5:0.0922719):0.202381,dipOrd1:0.210819):0.0258471,(hetGla2:0.0917322,cavPor3:0.136876):0.0994271):0.00910944,speTri2:0.145483):0.0274969,(oryCun2:0.109639,ochPri2:0.200966):0.102067):0.0141654,(((((((((hg19:0.00674057,panTro4:0.00692231):0.00309904,gorGor3:0.00918625):0.00954082,ponAbe2:0.0191843):0.00356049,nomLeu2:0.0218207):0.0116848,(rheMac3:0.00814945,papHam1:0.0079848):0.0289473):0.0208338,(calJac3:0.0342405,saiBol1:0.0333221):0.0359171):0.0594469,tarSyr1:0.137467):0.011091,(micMur1:0.0918138,otoGar3:0.127231):0.0351527):0.0153171,tupBel1:0.18879):0.0042463):0.0214646,((susScr3:0.121641,(vicPac1:0.109818,(turTru2:0.0635753,(oviAri1:0.0392493,bosTau7:0.0315816):0.0939861):0.0203711):0.00368417):0.0444758,((((felCat5:0.0897448,(canFam3:0.0888602,ailMel1:0.0767935):0.021837):0.05011,equCab2:0.109367):0.00605998,(myoLuc2:0.137144,pteVam1:0.114013):0.0339604):0.00395001,(eriEur1:0.226934,sorAra1:0.270619):0.0628319):0.00292667):0.0291403):0.0231397,((((loxAfr3:0.078841,proCap1:0.160295):0.00825096,echTel1:0.266786):0.0031636,triMan1:0.0685675):0.0736043,(dasNov3:0.112086,choHof1:0.0974658):0.0535724):0.00739115):0.245967,(monDom5:0.139913,(sarHar1:0.132596,macEug2:0.111778):0.0294309):0.21273):0.0770867,ornAna1:0.50425):0.135096,(((((melGal1:0.067697,galGal4:0.05253):0.13729,taeGut1:0.202681):0.00899388,melUnd1:0.127774):0.216078,anoCar2:0.575186):0.0128221,chrPic1:0.201659):0.137011):0.113527,xenTro3:0.943162):0.0646458,latCha1:0.596956):0.463611,((((((tetNig2:0.223213,fr3:0.198755):0.263107,oreNil2:0.33649):0.0139699,gasAcu1:0.314841):0.0573697,oryLat2:0.430105):0.185668,gadMor1:0.562778):0.169352,danRer7:0.753326):0.117017):0.501088,petMar1:0.501088);

    #   four different subset lists:
    paste glire.list euarchontoglires.list placental.list all.list
# mm10    mm10    mm10    mm10
# rn5     rn5     rn5     rn5
# dipOrd1 dipOrd1 dipOrd1 dipOrd1
# hetGla2 hetGla2 hetGla2 hetGla2
# cavPor3 cavPor3 cavPor3 cavPor3
# speTri2 speTri2 speTri2 speTri2
# oryCun2 oryCun2 oryCun2 oryCun2
# ochPri2 ochPri2 ochPri2 ochPri2
#         tupBel1 tupBel1 tupBel1
#         hg19    hg19    hg19
#         gorGor3 gorGor3 gorGor3
#         panTro4 panTro4 panTro4
#         nomLeu2 nomLeu2 nomLeu2
#         ponAbe2 ponAbe2 ponAbe2
#         tarSyr1 tarSyr1 tarSyr1
#         rheMac3 rheMac3 rheMac3
#         papHam1 papHam1 papHam1
#         otoGar3 otoGar3 otoGar3
#         calJac3 calJac3 calJac3
#         micMur1 micMur1 micMur1
#         saiBol1 saiBol1 saiBol1
#                 equCab2 equCab2
#                 vicPac1 vicPac1
#                 turTru2 turTru2
#                 susScr3 susScr3
#                 bosTau7 bosTau7
#                 oviAri1 oviAri1
#                 pteVam1 pteVam1
#                 myoLuc2 myoLuc2
#                 felCat5 felCat5
#                 canFam3 canFam3
#                 ailMel1 ailMel1
#                 eriEur1 eriEur1
#                 sorAra1 sorAra1
#                 choHof1 choHof1
#                 dasNov3 dasNov3
#                 proCap1 proCap1
#                 echTel1 echTel1
#                 triMan1 triMan1
#                 loxAfr3 loxAfr3
#                         macEug2
#                         sarHar1
#                         monDom5
#                         ornAna1
#                         galGal4
#                         taeGut1
#                         melGal1
#                         melUnd1
#                         anoCar2
#                         chrPic1
#                         xenTro3
#                         latCha1
#                         gadMor1
#                         gasAcu1
#                         fr3
#                         oreNil2
#                         tetNig2
#                         danRer7
#                         oryLat2
#                         petMar1

    # on organisms that do not have all species in all files, the file names
    #	need to be filtered.  Using this perl script to extract from
    # the full mfa files, only the subset of species from the four lists:
    cat << '_EOF_' > filterMfa.pl
#!/usr/bin/env perl

use strict;
use warnings;

my $argc = scalar(@ARGV);

if ($argc != 1) {
    printf STDERR "usage: filterMfa.pl <subset.list>\n";
    exit 255;
}

my %dbList;
my $file = shift;
open (FH, "<$file") or die "can not read $file";
printf STDERR "using list: $file\n";
while (my $db = <FH>) {
    chomp $db;
    $dbList{$db} = 1;
}
close (FH);

my $dirName = $file;
$dirName =~ s/.list//;
$dirName .= "Mfa";

my @mfaFileList = split('\n', `ls mfa/*.mfa`);
for (my $i = 0; $i < scalar(@mfaFileList); ++$i) {
    my $file = $mfaFileList[$i];
    my $chr = $file;
    $chr =~ s#^mfa/##;
#    printf STDERR "processing: %s into %s/%s\n", $file, $dirName, $chr;
    open (FH, "<$file") or die "can not read $file";
    open (OF, ">$dirName/$chr") or die "can not write to $dirName/$chr";
    my $inGroup = 0;
    while (my $line = <FH>) {
        if ($line =~ m/^> /) {
          chomp $line;
          my ($faHead, $faDbName) = split('\s+', $line);
          if (exists($dbList{$faDbName})) {
              $inGroup = 1;
                printf OF "> %s\n", $faDbName;
          } else {
              $inGroup = 0;
          }
        } elsif ($inGroup) {
            printf OF "%s", $line;
        }
    }
    close (FH);
    close (OF);
}
'_EOF_'
    # << happy emacs
    chmod +x filterMfa.pl

    mkdir glireMfa euarchontogliresMfa placentalMfa vertebrateMfa

    # extract each set from the full mfa files, run msa_view on
    #   each subset and construct .nh tree for that subset
    for N in glire euarchontoglires placental vertebrate
do
    ./filterMfa.pl ${N}.list
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/msa_view \
	--aggregate "`cat ${N}.list|xargs echo`" ${N}Mfa/*.mfa \
        | sed s/"> "/">"/ > 4d.${N}.mfa
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/tree_doctor \
        --no-branchlen --prune-all-but="`cat ${N}.list|xargs echo`" \
        ../tree-commas.nh > tree-commas.${N}.nh
done

    ### XXX ### MOST INTERESTING, this phyloFit operation was repeated
    ### to verify that the full 60 species vertebrate operation produced the
    ### same result as the original "all" subset.  This phyloFit appears to
    ### produce a different result each time ?
    # use phyloFit to create tree model (output is phyloFit.mod)
    for N in glire euarchontoglires placental vertebrate
do
    time nice -n +19 \
	/cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/phyloFit \
	    --EM --precision MED --msa-format FASTA --subst-mod REV \
		--tree ./tree-commas.${N}.nh 4d.${N}.mfa
    mv phyloFit.mod ${N}.mod
    grep TREE ${N}.mod | sed 's/TREE\:\ //' > ${N}.Nway.nh
done
    #   real    0m15.747s
    #   real    4m5.526s
    #   real    20m45.982s
    #   real    141m21.248s

#######################################################################
# phastCons 60-way (DONE - 2012-06-12, 2012-08-21 - Hiram)
    #	was unable to split the full chrom MAF files, now working on the
    #	maf files as they were split up during multiz

    # split 60way mafs into 10M chunks and generate sufficient statistics
    # files for # phastCons
    ssh encodek
    mkdir -p /hive/data/genomes/mm10/bed/multiz60way/cons/ss
    mkdir -p /hive/data/genomes/mm10/bed/multiz60way/cons/msa.split
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/msa.split

    cat << '_EOF_' > doSplit.csh
#!/bin/csh -ef
set c = $1
set MAF = /hive/data/genomes/mm10/bed/multiz60way/anno/result/$c.maf
set WINDOWS = /hive/data/genomes/mm10/bed/multiz60way/cons/ss/$c
set WC = `cat $MAF | wc -l`
set NL = `grep "^#" $MAF | wc -l`
if ( -s $2 ) then
    exit 0
endif
if ( -s $2.running ) then
    exit 0
endif

date >> $2.running

rm -fr $WINDOWS
mkdir $WINDOWS
pushd $WINDOWS > /dev/null
if ( $WC != $NL ) then
/cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/msa_split \
    $MAF -i MAF -o SS -r $WINDOWS/$c -w 10000000,0 -I 1000 -B 5000
endif
popd > /dev/null
date >> $2
rm -f $2.running
'_EOF_'
    # << happy emacs
    chmod +x doSplit.csh

    cat << '_EOF_' > template
#LOOP
doSplit.csh $(root1) {check out line+ $(root1).done}
#ENDLOOP
'_EOF_'
    # << happy emacs

    #	do the easy ones first to see some immediate results
    ls -1S -r ../../anno/result | sed -e "s/.maf//;" > maf.list

    gensub2 maf.list single template jobList
    para -ram=8g create jobList
    para try ... check ... etc
# Completed: 64 of 66 jobs
# Crashed: 2 jobs
# CPU time in finished jobs:     347730s    5795.49m    96.59h    4.02d  0.011 y
# IO & Wait Time:                102813s    1713.56m    28.56h    1.19d  0.003 y
# Average job time:                7040s     117.33m     1.96h    0.08d
# Longest finished job:           42666s     711.10m    11.85h    0.49d
# Submission to last job:        150336s    2505.60m    41.76h    1.74d
    # finish the last two on hgwdev with more memory.
# linux data memory, in 1024-byte units
export M=188743680
ulimit -S -m $M -v $M
./doSplit.csh chr1 chr1.done &
./doSplit.csh chr2 chr2.done
wait
    #   real    864m53.235s

    # Run phastCons
    #	This job is I/O intensive in its output files, beware where this
    #	takes place or do not run too many at once.
    ssh swarm
    mkdir -p /hive/data/genomes/mm10/bed/multiz60way/cons/run.cons
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/run.cons

    #	there are going to be several different phastCons runs using
    #	this same script.  They trigger off of the current working directory
    #	$cwd:t which is the "grp" in this script.  It is one of:
    #	all glire glirePrimate glirePrimatePlacental

    cat << '_EOF_' > doPhast.csh
#!/bin/csh -fe
set PHASTBIN = /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin
set c = $1
set f = $2
set len = $3
set cov = $4
set rho = $5
set grp = $cwd:t
set cons = /hive/data/genomes/mm10/bed/multiz60way/cons
set tmp = $cons/tmp/$f
mkdir -p $tmp
set ssSrc = $cons/ss
set useGrp = "$grp.mod"
if (-s $cons/$grp/$grp.non-inf) then
  ln -s $cons/$grp/$grp.mod $tmp
  ln -s $cons/$grp/$grp.non-inf $tmp
  ln -s $ssSrc/$c/$f.ss $tmp
else
  ln -s $ssSrc/$c/$f.ss $tmp
  ln -s $cons/$grp/$grp.mod $tmp
endif
pushd $tmp > /dev/null
if (-s $grp.non-inf) then
  $PHASTBIN/phastCons $f.ss $useGrp \
    --rho $rho --expected-length $len --target-coverage $cov --quiet \
    --not-informative `cat $grp.non-inf` \
    --seqname $c --idpref $c --most-conserved $f.bed --score > $f.pp
else
  $PHASTBIN/phastCons $f.ss $useGrp \
    --rho $rho --expected-length $len --target-coverage $cov --quiet \
    --seqname $c --idpref $c --most-conserved $f.bed --score > $f.pp
endif
popd > /dev/null
mkdir -p pp/$c bed/$c
sleep 4
touch pp/$c bed/$c
rm -f pp/$c/$f.pp
rm -f bed/$c/$f.bed
mv $tmp/$f.pp pp/$c
mv $tmp/$f.bed bed/$c
rm -fr $tmp
'_EOF_'
    # << happy emacs
    chmod a+x doPhast.csh

    #	this template will serve for all runs
    #	root1 == chrom name, file1 == ss file name without .ss suffix
    cat << '_EOF_' > template
#LOOP
../run.cons/doPhast.csh $(root1) $(file1) 45 0.3 0.3 {check out line+ pp/$(root1)/$(file1).pp}
#ENDLOOP
'_EOF_'
    # << happy emacs

    ls -1S ../ss/chr*/chr* | sed -e "s/.ss$//" > ss.list

    # Create parasol batch and run it
    ############################ run for all species
    cd /hive/data/genomes/mm10/bed/multiz60way/cons
    mkdir all
    cd all
    cp -p ../../4d/all.mod ./all.mod

    gensub2 ../run.cons/ss.list single ../run.cons/template jobList
    para -ram=8g create jobList
    para try ... check ... push ... etc.
# Completed: 314 of 314 jobs
# CPU time in finished jobs:      36286s     604.77m    10.08h    0.42d  0.001 y
# IO & Wait Time:                 10101s     168.35m     2.81h    0.12d  0.000 y
# Average job time:                 148s       2.46m     0.04h    0.00d
# Longest finished job:             219s       3.65m     0.06h    0.00d
# Submission to last job:          4383s      73.05m     1.22h    0.05d

    # create Most Conserved track
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/all
    cut -f1 ../../../../chrom.sizes | while read C
do
    ls -d bed/${C} 2> /dev/null | while read D
    do
        cat ${D}/${C}*.bed
    done | sort -k1,1 -k2,2n \
    | awk '{printf "%s\t%d\t%d\tlod=%d\t%s\n", "'${C}'", $2, $3, $5, $5;}'
done > tmpMostConserved.bed
    #   -rw-rw-r--  1 230642249 Jun 15 11:48 tmpMostConserved.bed
    /cluster/bin/scripts/lodToBedScore tmpMostConserved.bed > mostConserved.bed
    #   -rw-rw-r--  1 236425914 Jun 15 11:52 mostConserved.bed

    # load into database
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/all
    time nice -n +19 hgLoadBed mm10 phastConsElements60way mostConserved.bed
    #   Read 6748481 elements of size 5 from mostConserved.bed
    #   real    2m20.950s

    # Try for 5% overall cov, and 70% CDS cov
    featureBits mm10 -enrichment refGene:cds phastConsElements60way
    #	--rho 0.3 --expected-length 45 --target-coverage 0.3
    #   refGene:cds 1.281%, phastConsElements60way 6.517%,
    #   both 0.913%, cover 71.29%, enrich 10.94x
    time featureBits mm10 -enrichment ensGene:cds phastConsElements60way
    #   ensGene:cds 1.357%, phastConsElements60way 6.517%, both 0.942%, cover
    #   69.39%, enrich 10.65x
    #   real    0m54.109s
    time featureBits mm10 -enrichment knownGene:cds phastConsElements60way
    #   knownGene:cds 1.325%, phastConsElements60way 6.517%, both 0.930%,
    #   cover 70.18%, enrich 10.77x
    #   real    0m50.472s

    # Create merged posterier probability file and wiggle track data files
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/all
    mkdir downloads

    for D in `ls -d pp/chr* | sed -e 's#pp/##'`
do
    echo "working: $D"
    find ./pp/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.wigFix.gz
done
    #   real    102m58.496s

    #	encode those files into wiggle data
    time (zcat downloads/*.wigFix.gz \
	| wigEncode stdin phastCons60way.wig phastCons60way.wib)
    #   Converted stdin, upper limit 1.00, lower limit 0.00
    #   real    9m32.980s
    du -hsc *.wi?
    #   1.8G    phastCons60way.wib
    #   298M    phastCons60way.wig
    #   2.1G    total

    #	encode into a bigWig file:
    #	(warning wigToBigWig process grows to about 36 Gb)
    #	in bash, to avoid the 32 Gb memory limit, set 180 Gb here:
sizeG=188743680
export sizeG
ulimit -d $sizeG
ulimit -v $sizeG
    time (zcat downloads/*.wigFix.gz \
        | wigToBigWig stdin ../../../../chrom.sizes phastCons60way.bw)
    #   real    27m1.039s
    #   -rw-rw-r--  1 4671685725 Jun 18 10:24 phastCons60way.bw
    bigWigInfo phastCons60way.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 3,333,510,917
primaryIndexSize: 100,774,056
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.149660
min: 0.000000
max: 1.000000
std: 0.282516

    #	if you wanted to use the bigWig file, loading bigWig table:
    #   but we don't use the bigWig file
    mkdir /gbdb/mm10/bbi
    ln -s `pwd`/phastCons60way.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phastCons60way; \
            create table phastCons60way (fileName varchar(255) not null); \
            insert into phastCons60way values
	("/gbdb/mm10/bbi/phastCons60way.bw");'

    # Load gbdb and database with wiggle.
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/all
    ln -s `pwd`/phastCons60way.wib /gbdb/mm10/multiz60way/phastCons60way.wib
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phastCons60way phastCons60way.wig
    #   real    0m54.546s

    wigTableStats.sh mm10 phastCons60way
# db.table      min max mean count sumData
# mm10.phastCons60way     0 1 0.14966 1929686275 2.88797e+08
#       stdDev viewLimits
#       0.282516 viewLimits=0:1

    #  Create histogram to get an overview of all the data
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/all
    time nice -n +19 hgWiggle -doHistogram -db=mm10 \
	-hBinSize=0.001 -hBinCount=1000 -hMinVal=0.0 -verbose=2 \
	    phastCons60way > histogram.data 2>&1
    #	real    7m37.212s

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phastCons60way track"
set xlabel " phastCons60way score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.02]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ########################################################################
    ### Create a phastCons data set for Glires

    # setup glire-only run
    ssh swarm
    mkdir /hive/data/genomes/mm10/bed/multiz60way/cons/glire
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/glire
    # glire-only: get the glire only tree from the 4d directory
    cp -p ../../4d/glire.mod ./glire.mod
    # and all the others become the non-informative list for phastCons to ignore
    sort ../../4d/glire.list > glire.list
    sort ../../4d/vertebrate.list > vertebrate.list
    comm -13 glire.list vertebrate.list | xargs echo \
        | sed -e "s/ /,/g" > glire.non-inf

    gensub2 ../run.cons/ss.list single ../run.cons/template jobList
    para -ram=8g create jobList
    para try ... check ... push ... etc.
# Completed: 314 of 314 jobs
# CPU time in finished jobs:      12411s     206.85m     3.45h    0.14d  0.000 y
# IO & Wait Time:                117850s    1964.16m    32.74h    1.36d  0.004 y
# Average job time:                 415s       6.91m     0.12h    0.00d
# Longest finished job:             658s      10.97m     0.18h    0.01d
# Submission to last job:           796s      13.27m     0.22h    0.01d

    cd /hive/data/genomes/mm10/bed/multiz60way/cons/glire
    # create Most Conserved track
    cut -f1 ../../../../chrom.sizes | while read C
do
    ls -d bed/${C} 2> /dev/null | while read D
    do
        cat ${D}/${C}*.bed
    done | sort -k1,1 -k2,2n \
    | awk '{printf "%s\t%d\t%d\tlod=%d\t%s\n", "'${C}'", $2, $3, $5, $5;}'
done > tmpMostConserved.bed
    #   real    0m32.945s
    /cluster/bin/scripts/lodToBedScore tmpMostConserved.bed > mostConserved.bed
    #   real    0m19.122s

    featureBits mm10 mostConserved.bed
    #   117058023 bases of 2652783500 (4.413%) in intersection
    #   real    0m21.506s

    # load into database
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/glire
    time nice -n +19 hgLoadBed mm10 phastConsElements60wayGlire \
	mostConserved.bed
    #	Loaded 1336504 elements of size 6
    #	real    0m13.672s
    # verify coverage
    time featureBits mm10 phastConsElements60wayGlire
    #   117058023 bases of 2652783500 (4.413%) in intersection
    #   real    0m15.041s

    #	--rho 0.3 --expected-length 45 --target-coverage 0.3
    featureBits mm10 -enrichment refGene:cds phastConsElements60wayGlire
    #   refGene:cds 1.282%, phastConsElements60wayGlire 4.413%,
    #   both 0.944%, cover 73.60%, enrich 16.68x

    featureBits mm10 -enrichment knownGene:cds phastConsElements60wayGlire
    #   knownGene:cds 1.325%, phastConsElements60wayGlire 4.413%,
    #   both 0.957%, cover 72.22%, enrich 16.37x

#	Create the downloads .pp files, from which the phastCons wiggle data
    #	is calculated
    # sort by chromName, chromStart so that items are in numerical order
    #  for wigEncode
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/glire
    mkdir downloads
    for D in `ls -d pp/chr* | sed -e 's#pp/##'`
do
    echo "working: $D"
    find ./pp/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.glire.wigFix.gz
done

    # Create merged posterier probability file and wiggle track data files
    time (zcat downloads/chr*.wigFix.gz \
	 | wigEncode stdin phastCons60wayGlire.wig phastCons60wayGlire.wib) &
    #   Converted stdin, upper limit 1.00, lower limit 0.00
    #   real    10m26.712s

    #	encode to bigWig
    #	(warning wigToBigWig process grows to about 36 Gb)
    #	in bash, to avoid the 32 Gb memory limit:
export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG

    time (zcat downloads/*.wigFix.gz \
        | wigToBigWig -verbose=2 stdin ../../../../chrom.sizes \
        phastCons60wayGlire.bw > bigWig.log 2>&1) &
    #   real    52m17.108s
    grep VmPeak bigWig.log
    # pid=5552: VmPeak:     20926360 kB

    bigWigInfo phastCons60wayGlire.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 3,631,413,425
primaryIndexSize: 100,774,056
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.142675
min: 0.000000
max: 1.000000
std: 0.252347

    #	if desired to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phastCons60wayGlire.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phastCons60wayGlire; \
            create table phastCons60wayGlire \
		(fileName varchar(255) not null); \
            insert into phastCons60wayGlire values
	("/gbdb/mm10/bbi/phastCons60wayGlire.bw");'

    ## load table with wiggle data
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/glire
    ln -s `pwd`/phastCons60wayGlire.wib \
	/gbdb/mm10/multiz60way/phastCons60wayGlire.wib
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phastCons60wayGlire phastCons60wayGlire.wig
    #   real    0m56.786s

    wigTableStats.sh mm10 phastCons60wayGlire
# db.table      min max mean count sumData
mm10.phastCons60wayGlire     0 1 0.142675 1929686275 2.75318e+08
#	stdDev viewLimits
#       0.252347 viewLimits=0:1

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.001 -hBinCount=1000 -hMinVal=0.0 -verbose=2 \
	    -db=mm10 phastCons60wayGlire  > histogram.data 2>&1
    #	real    4m28.743s

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Hg19 Histogram phastCons60wayGlire track"
set xlabel " phastCons60wayGlire score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.02]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ########################################################################
    ### Create a phastCons data set for Euarchontoglires

    # setup euarchontoglires-only run
    ssh swarm
    mkdir /hive/data/genomes/mm10/bed/multiz60way/cons/euarchontoglires
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/euarchontoglires
    # euarchontoglires-only: get the euarchontoglires only tree from the 4d directory
    cp -p ../../4d/euarchontoglires.mod ./euarchontoglires.mod
    # and all the others become the non-informative list for phastCons to ignore
    sort ../../4d/euarchontoglires.list > euarchontoglires.list
    sort ../../4d/vertebrate.list > vertebrate.list
    comm -13 euarchontoglires.list vertebrate.list | xargs echo \
        | sed -e "s/ /,/g" > euarchontoglires.non-inf

    gensub2 ../run.cons/ss.list single ../run.cons/template jobList
    para -ram=8g create jobList
    para try ... check ... push ... etc.
# Completed: 314 of 314 jobs
# CPU time in finished jobs:      17421s     290.36m     4.84h    0.20d  0.001 y
# IO & Wait Time:                 37430s     623.83m    10.40h    0.43d  0.001 y
# Average job time:                 175s       2.91m     0.05h    0.00d
# Longest finished job:             343s       5.72m     0.10h    0.00d
# Submission to last job:          2403s      40.05m     0.67h    0.03d

    cd /hive/data/genomes/mm10/bed/multiz60way/cons/euarchontoglires
    # create Most Conserved track
    cut -f1 ../../../../chrom.sizes | while read C
do
    ls -d bed/${C} 2> /dev/null | while read D
    do
        cat ${D}/${C}*.bed
    done | sort -k1,1 -k2,2n \
    | awk '{printf "%s\t%d\t%d\tlod=%d\t%s\n", "'${C}'", $2, $3, $5, $5;}'
done > tmpMostConserved.bed
    #   real    0m32.945s
    /cluster/bin/scripts/lodToBedScore tmpMostConserved.bed > mostConserved.bed
    #   real    0m19.122s

    featureBits mm10 mostConserved.bed
    #   127113541 bases of 2652783500 (4.792%) in intersection
    #   real    0m21.506s

    # load into database
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/euarchontoglires
    time nice -n +19 hgLoadBed mm10 phastConsElements60wayEuarchontoGlires \
	mostConserved.bed
    #	Loaded 2327130 elements of size 6
    #	real    0m24.591s
    # verify coverage
    time featureBits mm10 phastConsElements60wayEuarchontoGlires
    #   127113541 bases of 2652783500 (4.792%) in intersection
    #   real    0m18.857s

    #	--rho 0.3 --expected-length 45 --target-coverage 0.3
    featureBits mm10 -enrichment refGene:cds phastConsElements60wayEuarchontoGlires
    #   refGene:cds 1.282%, phastConsElements60wayEuarchontoGlires 4.792%,
    #   both 0.929%, cover 72.46%, enrich 15.12x

    featureBits mm10 -enrichment knownGene:cds phastConsElements60wayEuarchontoGlires
    #   knownGene:cds 1.325%, phastConsElements60wayEuarchontoGlires 4.792%,
    #   both 0.943%, cover 71.16%, enrich 14.85x

    #	Create the downloads .pp files, from which the phastCons wiggle data
    #	is calculated
    # sort by chromName, chromStart so that items are in numerical order
    #  for wigEncode
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/euarchontoglires
    mkdir downloads
    for D in `ls -d pp/chr* | sed -e 's#pp/##'`
do
    echo "working: $D"
    find ./pp/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.euarchontoglires.wigFix.gz
done

    # Create merged posterier probability file and wiggle track data files
    time (zcat downloads/chr*.wigFix.gz \
	 | wigEncode stdin phastCons60wayEuarchontoGlires.wig phastCons60wayEuarchontoGlires.wib \
        > wigEncode.log 2>&1) &
    # Converted stdin, upper limit 1.00, lower limit 0.00
    #   real    9m49.080s

    #	encode to bigWig
    #	(warning wigToBigWig process grows to about 36 Gb)
    #	in bash, to avoid the 32 Gb memory limit:
export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG

    time (zcat downloads/*.wigFix.gz \
        | wigToBigWig stdin ../../../../chrom.sizes phastCons60wayEuarchontoGlires.bw \
        > bigWig.log 2>&1 ) &
    #   real    26m0.111s
    bigWigInfo phastCons60wayEuarchontoGlires.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 3,411,704,465
primaryIndexSize: 100,774,056
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.133253
min: 0.000000
max: 1.000000
std: 0.256320

    #	if desired to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phastCons60wayEuarchontoGlires.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phastCons60wayEuarchontoGlires; \
            create table phastCons60wayEuarchontoGlires \
		(fileName varchar(255) not null); \
            insert into phastCons60wayEuarchontoGlires values
	("/gbdb/mm10/bbi/phastCons60wayEuarchontoGlires.bw");'

    ## load table with wiggle data
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/euarchontoglires
    ln -s `pwd`/phastCons60wayEuarchontoGlires.wib \
	/gbdb/mm10/multiz60way/phastCons60wayEuarchontoGlires.wib
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phastCons60wayEuarchontoGlires phastCons60wayEuarchontoGlires.wig
    #   real    0m50.676s

    time wigTableStats.sh mm10 phastCons60wayEuarchontoGlires
# db.table      min max mean count sumData
mm10.phastCons60wayEuarchontoGlires  0 1 0.133253 1929686275 2.57137e+08
#	stdDev viewLimits
#       0.25632 viewLimits=0:1
    #   real    0m21.964s

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.001 -hBinCount=1000 -hMinVal=0.0 -verbose=2 \
	    -db=mm10 phastCons60wayEuarchontoGlires  > histogram.data 2>&1
    #	real    3m31.112s

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phastCons60wayEuarchontoGlires track"
set xlabel " phastCons60wayEuarchontoGlires score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.02]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ########################################################################
    ### Create a phastCons data set for primate ***### This was constructed
    ### and examined, but not used in the release

    # setup primate-only run
    ssh swarm
    mkdir /hive/data/genomes/mm10/bed/multiz60way/cons/primate
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/primate
    # primate-only: get the primate only tree from the 4d directory
    cp -p ../../4d/primate.mod ./primate.mod
    # and all the others become the non-informative list for phastCons to ignore
    cat ../../4d/glire.list ../../4d/placental.list ../../4d/vertebrate.list \
        | grep -v mm10 | sort | xargs echo | sed -e "s/ /,/g" \
        > primate.non-inf

    gensub2 ../run.cons/ss.list single ../run.cons/template jobList
    para -ram=8g create jobList
    para try ... check ... push ... etc.
# Completed: 314 of 314 jobs
# CPU time in finished jobs:      13884s     231.39m     3.86h    0.16d  0.000 y
# IO & Wait Time:                130791s    2179.86m    36.33h    1.51d  0.004 y
# Average job time:                 461s       7.68m     0.13h    0.01d
# Longest finished job:             741s      12.35m     0.21h    0.01d
# Submission to last job:           910s      15.17m     0.25h    0.01d

    cd /hive/data/genomes/mm10/bed/multiz60way/cons/primate
    # create Most Conserved track
    cut -f1 ../../../../chrom.sizes | while read C
do
    ls -d bed/${C} 2> /dev/null | while read D
    do
        cat ${D}/${C}*.bed
    done | sort -k1,1 -k2,2n \
    | awk '{printf "%s\t%d\t%d\tlod=%d\t%s\n", "'${C}'", $2, $3, $5, $5;}'
done > tmpMostConserved.bed

    /cluster/bin/scripts/lodToBedScore tmpMostConserved.bed > mostConserved.bed
    #   real    0m27.199s

    featureBits mm10 mostConserved.bed
    #   112908553 bases of 2652783500 (4.256%) in intersection

    # load into database
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/primate
    time nice -n +19 hgLoadBed mm10 phastConsElements60wayPrimate \
	mostConserved.bed
    #	Loaded 1119924 elements of size 6
    #	real    0m17.423s
    # verify coverage
    featureBits mm10 phastConsElements60wayPrimate
    #   112908553 bases of 2652783500 (4.256%) in intersection
    #   real    0m13.684s

    #	--rho 0.3 --expected-length 45 --target-coverage 0.3
    featureBits mm10 -enrichment refGene:cds phastConsElements60wayPrimate
    #   refGene:cds 1.281%, phastConsElements60wayPrimate 4.256%,
    #   both 0.897%, cover 69.98%, enrich 16.44x

    featureBits mm10 -enrichment knownGene:cds phastConsElements60wayPrimate
    #   knownGene:cds 1.325%, phastConsElements60wayPrimate 4.256%,
    #   both 0.909%, cover 68.64%, enrich 16.13x

    featureBits mm10 -enrichment ensGene:cds phastConsElements60wayPrimate
    #   ensGene:cds 1.357%, phastConsElements60wayPrimate 4.256%, both 0.913%,
    #   cover 67.30%, enrich 15.81x

    #	Create the downloads .pp files, from which the phastCons wiggle data
    #	is calculated
    # sort by chromName, chromStart so that items are in numerical order
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/primate
    mkdir downloads
    for D in `ls -d pp/chr* | sed -e 's#pp/##'`
do
    echo "working: $D"
    find ./pp/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.primate.wigFix.gz
done

    # Create merged posterier probability file and wiggle track data files
    zcat downloads/chr*.wigFix.gz \
	 | wigEncode stdin phastCons60wayPrimate.wig phastCons60wayPrimate.wib
    # Converted stdin, upper limit 1.00, lower limit 0.00
    #	real    12m22.465s

    #	encode to bigWig
    #	(warning wigToBigWig process grows to about 36 Gb)
    #	in bash, to avoid the 32 Gb memory limit:
export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG

    zcat downloads/*.wigFix.gz \
        | wigToBigWig stdin ../../../../chrom.sizes phastCons60wayPrimate.bw
    #    real 31m44.517s
    bigWigInfo phastCons60wayPrimate.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 2,431,379,060
primaryIndexSize: 100,774,056
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.093847
min: 0.000000
max: 1.000000
std: 0.233892

    #	if desired to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phastCons60wayPrimate.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phastCons60wayPrimate; \
            create table phastCons60wayPrimate \
		(fileName varchar(255) not null); \
            insert into phastCons60wayPrimate values
	("/gbdb/mm10/bbi/phastCons60wayPrimate.bw");'

    ## load table with wiggle data
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/primate
    ln -s `pwd`/phastCons60wayPrimate.wib \
	/gbdb/mm10/multiz60way/phastCons60wayPrimate.wib
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phastCons60wayPrimate phastCons60wayPrimate.wig
    #   real    1m24.188s

    wigTableStats.sh mm10 phastCons60wayPrimate
# db.table      min max mean count sumData
# mm10.phastCons60wayPrimate 0 1 0.0938475 1929686275 1.81096e+08
#       0.233892 viewLimits=0:1

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.001 -hBinCount=1000 -hMinVal=0.0 -verbose=2 \
	    -db=mm10 phastCons60wayPrimate  > histogram.data 2>&1
    #   real    7m3.198s

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phastCons60wayPrimate track"
set xlabel " phastCons60wayPrimate score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.02]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

#########################################################################
    ### Create a phastCons data set for Placental

    # setup placental-only run
    ssh swarm
    mkdir /hive/data/genomes/mm10/bed/multiz60way/cons/placental
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/placental

    # placental-only: get the placental only tree from the 4d directory
    cp -p ../../4d/placental.mod ./placental.mod
    # and all the others become the non-informative list for phastCons to ignore
    sort ../../4d/placental.list > placental.list
    sort ../../4d/vertebrate.list > vertebrate.list
    comm -13 placental.list vertebrate.list | xargs echo \
        | sed -e "s/ /,/g" > placental.non-inf

    gensub2 ../run.cons/ss.list single ../run.cons/template jobList
    para create jobList
    para try ... check ... push ... etc.
# Completed: 314 of 314 jobs
# CPU time in finished jobs:      27853s     464.21m     7.74h    0.32d  0.001 y
# IO & Wait Time:                128981s    2149.69m    35.83h    1.49d  0.004 y
# Average job time:                 499s       8.32m     0.14h    0.01d
# Longest finished job:             785s      13.08m     0.22h    0.01d
# Submission to last job:          5970s      99.50m     1.66h    0.07d

    cd /hive/data/genomes/mm10/bed/multiz60way/cons/placental
    # create Most Conserved track
    cut -f1 ../../../../chrom.sizes | while read C
do
    ls -d bed/${C} 2> /dev/null | while read D
    do
        cat ${D}/${C}*.bed
    done | sort -k1,1 -k2,2n \
    | awk '{printf "%s\t%d\t%d\tlod=%d\t%s\n", "'${C}'", $2, $3, $5, $5;}'
done > tmpMostConserved.bed
    #   real    0m44.506s
    /cluster/bin/scripts/lodToBedScore tmpMostConserved.bed > mostConserved.bed
    #   real    0m44.170s

    featureBits mm10 mostConserved.bed
    #   144041584 bases of 2652783500 (5.430%) in intersection
    #   real    0m54.927s

    # load into database
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/placental
    time nice -n +19 hgLoadBed mm10 phastConsElements60wayPlacental \
	mostConserved.bed
    #	Loaded 5257437 elements of size 6
    #	real    0m56.788s

    # verify coverage, should be the same as the file measured above
    time featureBits mm10 phastConsElements60wayPlacental
    #   144041584 bases of 2652783500 (5.430%) in intersection
    #   real    0m39.537s

    #	--rho 0.3 --expected-length 45 --target-coverage 0.3
    time featureBits mm10 -enrichment refGene:cds phastConsElements60wayPlacental
    #   refGene:cds 1.282%, phastConsElements60wayPlacental 5.430%,
    #   both 0.920%, cover 71.73%, enrich 13.21x
    #   real    0m39.833s

    time featureBits mm10 -enrichment knownGene:cds phastConsElements60wayPlacental
    #   knownGene:cds 1.325%, phastConsElements60wayPlacental 5.430%,
    #   both 0.934%, cover 70.47%, enrich 12.98x
    #   real    0m44.567s

    time featureBits mm10 -enrichment ensGene:cds phastConsElements60wayPlacental
    #   ensGene:cds 1.357%, phastConsElements60wayPlacental 5.430%,
    #   both 0.941%, cover 69.32%, enrich 12.77x
    #   real    0m43.093s

    #	Create the downloads .pp files, from which the phastCons wiggle data
    #	is calculated
    # sort by chromName, chromStart so that items are in numerical order
    #  for wigEncode
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/placental
    mkdir downloads
    for D in `ls -d pp/chr* | sed -e 's#pp/##'`
do
    echo "working: $D"
    find ./pp/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.placental.wigFix.gz
done

    # Create merged posterier probability file and wiggle track data files
    time (zcat downloads/chr*.wigFix.gz \
	 | wigEncode stdin phastCons60wayPlacental.wig \
        phastCons60wayPlacental.wib > wigEncode.log 2>&1) &
    # Converted stdin, upper limit 1.00, lower limit 0.00
    #   real    9m48.237s

    #	encode to bigWig
    #	(warning wigToBigWig process grows to about 36 Gb)
    #	in bash, to avoid the 32 Gb memory limit:
export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG

    time (zcat downloads/*.wigFix.gz \
        | wigToBigWig stdin ../../../../chrom.sizes \
        phastCons60wayPlacental.bw > bigWig.log 2>&1) &
    #   real    25m18.556s
    bigWigInfo phastCons60wayPlacental.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 3,271,676,156
primaryIndexSize: 100,774,056
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.135703
min: 0.000000
max: 1.000000
std: 0.266432

    #	if desired to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phastCons60wayPlacental.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phastCons60wayPlacental; \
            create table phastCons60wayPlacental \
		(fileName varchar(255) not null); \
            insert into phastCons60wayPlacental values
	("/gbdb/mm10/bbi/phastCons60wayPlacental.bw");'

    ## load table with wiggle data
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/placental
    ln -s `pwd`/phastCons60wayPlacental.wib \
	/gbdb/mm10/multiz60way/phastCons60wayPlacental.wib
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phastCons60wayPlacental phastCons60wayPlacental.wig
    #   real    0m41.999s

    time wigTableStats.sh mm10 phastCons60wayPlacental
# db.table      min max mean count sumData
# mm10.phastCons60wayPlacental 0 1 0.135703 1929686275 2.61864e+08
#	stdDev viewLimits
#       0.266432 # viewLimits=0:1
    #   real    0m21.723s


    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.001 -hBinCount=1000 -hMinVal=0.0 -verbose=2 \
	    -db=mm10 phastCons60wayPlacental  > histogram.data 2>&1
    #   real    2m39.659s

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phastCons60wayPlacental track"
set xlabel " phastCons60wayPlacental score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.02]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

#########################################################################
    ### Create a phastCons data set for Vertebrate

    # setup vertebrate-only run
    ssh swarm
    mkdir /hive/data/genomes/mm10/bed/multiz60way/cons/vertebrate
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/vertebrate

    # vertebrate-only: get the vertebrate only tree from the 4d directory
    cp -p ../../4d/vertebrate.mod ./vertebrate.mod
    # they are all in this one, no need for non-informative list

    gensub2 ../run.cons/ss.list single ../run.cons/template jobList
    para create jobList
    para try ... check ... push ... etc.
# Completed: 313 of 314 jobs
# Crashed: 1 jobs
# CPU time in finished jobs:      36058s     600.97m    10.02h    0.42d  0.001 y
# IO & Wait Time:                125496s    2091.59m    34.86h    1.45d  0.004 y
# Average job time:                 516s       8.60m     0.14h    0.01d
# Longest finished job:             912s      15.20m     0.25h    0.01d
# Submission to last job:          2681s      44.68m     0.74h    0.03d
    # the one failed job was completed manually on hgwdev

    cd /hive/data/genomes/mm10/bed/multiz60way/cons/vertebrate
    # create Most Conserved track
    cut -f1 ../../../../chrom.sizes | while read C
do
    ls -d bed/${C} 2> /dev/null | while read D
    do
        cat ${D}/${C}*.bed
    done | sort -k1,1 -k2,2n \
    | awk '{printf "%s\t%d\t%d\tlod=%d\t%s\n", "'${C}'", $2, $3, $5, $5;}'
done > tmpMostConserved.bed
    #   real    0m44.506s
    /cluster/bin/scripts/lodToBedScore tmpMostConserved.bed > mostConserved.bed

    time featureBits mm10 mostConserved.bed
    #   172842314 bases of 2652783500 (6.516%) in intersection
    #   real    1m23.298s

    # load into database
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/vertebrate
    time nice -n +19 hgLoadBed mm10 phastConsElements60wayVertebrate \
	mostConserved.bed
    #   Read 6747163 elements of size 5 from mostConserved.bed
    #   real    1m15.122s

    # verify coverage
    featureBits mm10 phastConsElements60wayVertebrate
    #   172842314 bases of 2652783500 (6.516%) in intersection

    #	--rho 0.3 --expected-length 45 --target-coverage 0.3
    featureBits mm10 -enrichment refGene:cds phastConsElements60wayVertebrate
    #   refGene:cds 1.282%, phastConsElements60wayVertebrate 6.516%,
    #   both 0.914%, cover 71.26%, enrich 10.94x

    time featureBits mm10 -enrichment ensGene:cds phastConsElements60wayVertebrate
    #   ensGene:cds 1.357%, phastConsElements60wayVertebrate 6.516%,
    #   both 0.942%, cover 69.39%, enrich 10.65x
    #   real    0m51.139s

    time featureBits mm10 -enrichment knownGene:cds phastConsElements60wayVertebrate
    #   knownGene:cds 1.325%, phastConsElements60wayVertebrate 6.516%,
    #   both 0.930%, cover 70.18%, enrich 10.77x
    #   real    0m51.545s

    #	Create the downloads .pp files, from which the phastCons wiggle data
    #	is calculated
    # sort by chromName, chromStart so that items are in numerical order
    #  for wigEncode
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/vertebrate
    mkdir downloads
    for D in `ls -d pp/chr* | sed -e 's#pp/##'`
do
    echo "working: $D"
    find ./pp/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.vertebrate.wigFix.gz
done

    # Create merged posterier probability file and wiggle track data files
    time (zcat downloads/chr*.wigFix.gz \
	 | wigEncode stdin phastCons60wayVertebrate.wig \
        phastCons60wayVertebrate.wib > wigEncode.log 2>&1 ) &
    # Converted stdin, upper limit 1.00, lower limit 0.00
    #   real    9m48.554s

    #	encode to bigWig
    #	(warning wigToBigWig process grows to about 36 Gb)
    #	in bash, to avoid the 32 Gb memory limit:
export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG

    time (zcat downloads/*.wigFix.gz \
        | wigToBigWig stdin ../../../../chrom.sizes \
        phastCons60wayVertebrate.bw > bigWig.log 2>&1) &
    #   real    25m8.630s

    bigWigInfo phastCons60wayVertebrate.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 3,333,348,984
primaryIndexSize: 100,774,056
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.149646
min: 0.000000
max: 1.000000
std: 0.282502

    #	if desired to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phastCons60wayVertebrate.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phastCons60wayVertebrate; \
            create table phastCons60wayVertebrate \
		(fileName varchar(255) not null); \
            insert into phastCons60wayVertebrate values
	("/gbdb/mm10/bbi/phastCons60wayVertebrate.bw");'

    ## load table with wiggle data
    ssh hgwdev
    cd /hive/data/genomes/mm10/bed/multiz60way/cons/vertebrate
    ln -s `pwd`/phastCons60wayVertebrate.wib \
	/gbdb/mm10/multiz60way/phastCons60wayVertebrate.wib
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phastCons60wayVertebrate phastCons60wayVertebrate.wig
    #   real    0m45.432s

    time wigTableStats.sh mm10 phastCons60wayVertebrate
# db.table      min max mean count sumData
# mm10.phastCons60wayVertebrate 0 1 0.149646 1929686275 2.8877e+08
#	stdDev viewLimits
#       0.282502 viewLimits=0:1
    #   real    0m22.224s

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.001 -hBinCount=1000 -hMinVal=0.0 -verbose=2 \
	    -db=mm10 phastCons60wayVertebrate  > histogram.data 2>&1
    #   real    2m52.041s

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phastCons60wayVertebrate track"
set xlabel " phastCons60wayVertebrate score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.02]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

#########################################################################
# phyloP conservation for 60-way (DONE - 2012-06-15 - 2012-08-21 - Hiram)
#
# Vertebrate, Glire, Primate, Placental
#
    # split SS files into 1M chunks, this business needs smaller files
    #   to complete

    # many of these jobs run too much memory to finish on a kluster node
    # can run all of this on hgwdev

    mkdir /hive/data/genomes/mm10/bed/multiz60way/consPhyloP
    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP
    mkdir ss run.split
    cd run.split

    cat << '_EOF_' > doSplit.csh
#!/bin/csh -ef
set c = $1
set MAF = /hive/data/genomes/mm10/bed/multiz60way/anno/result/$c.maf
set WINDOWS = /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/ss/$c
set WC = `cat $MAF | wc -l`
set NL = `grep "^#" $MAF | wc -l`
if ( -s $2 ) then
    exit 0
endif
if ( -s $2.running ) then
    exit 0
endif

date >> $2.running

rm -fr $WINDOWS
mkdir $WINDOWS
pushd $WINDOWS > /dev/null
if ( $WC != $NL ) then
/cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/msa_split \
    $MAF -i MAF -o SS -r $WINDOWS/$c -w 1000000,0 -I 1000 -B 5000
endif
popd > /dev/null
date >> $2
rm -f $2.running
'_EOF_'
# << happy emacs

    #	do the easy ones first to see some immediate results
    ls -1S -r ../../anno/result | sed -e "s/.maf//;" > maf.list

    cat << '_EOF_' > template
#LOOP
./doSplit.csh $(root1) $(root1).done
#ENDLOOP
'_EOF_'
# << happy emacs

    gensub2 maf.list single template jobList
    # copy the jobList to runEm.sh, edit to make all the commands run in
    #   the background, with wait statements every few commands to run
    #   a small number of these at once, no more than four at once with
    #   the large chroms, the small randoms can run a bunch at once, they
    #   finish quickly.
    time ./runEm.sh
    # about 11h30m

    # run phyloP with score=LRT
    ssh swarm
    cd /cluster/data/mm10/bed/multiz60way/consPhyloP
    mkdir run.phyloP
    cd run.phyloP

    # Adjust model file base composition background and rate matrix to be
    # representative of the chromosomes in play
    grep BACKGROUND ../../cons/all/all.mod | awk '{printf "%0.3f\n", $3 + $4}'
    #	0.525
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/modFreqs \
	../../cons/all/all.mod 0.525 > all.mod
    grep BACKGROUND ../../cons/glire/glire.mod \
	| awk '{printf "%0.3f\n", $3 + $4}'
    #	0.531
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/modFreqs \
	../../cons/glire/glire.mod 0.531 > glire.mod
    grep BACKGROUND ../../cons/primate/primate.mod \
	| awk '{printf "%0.3f\n", $3 + $4}'
    #	0.509
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/modFreqs \
	../../cons/primate/primate.mod 0.509 > primate.mod
    grep BACKGROUND ../../cons/euarchontoglires/euarchontoglires.mod \
	| awk '{printf "%0.3f\n", $3 + $4}'
    #	0.518
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/modFreqs \
	../../cons/euarchontoglires/euarchontoglires.mod 0.518 \
        > euarchontoglires.mod

    grep BACKGROUND ../../cons/placental/placental.mod \
	| awk '{printf "%0.3f\n", $3 + $4}'
    #	0.525
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/modFreqs \
	../../cons/placental/placental.mod 0.525 > placental.mod
    grep BACKGROUND ../../cons/vertebrate/vertebrate.mod \
	| awk '{printf "%0.3f\n", $3 + $4}'
    #	0.525
    /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/modFreqs \
	../../cons/vertebrate/vertebrate.mod 0.525 > vertebrate.mod

    cat << '_EOF_' > doPhyloP.csh
#!/bin/csh -fex
set PHASTBIN = /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin
set f = $1
set ssFile = $1:t
echo "ssFile: $ssFile"
set out = $2
set cName = $f:h
echo "cName: $cName"
set n = $f:r:e
set grp = $cwd:t
set cons = /hive/data/genomes/mm10/bed/multiz60way/consPhyloP
set tmp = $cons/tmp/$grp/$f
rm -fr $tmp
mkdir -p $tmp
set ssSrc = "$cons/ss/$cName/$ssFile"
set useGrp = "$grp.mod"
ln -s $cons/run.phyloP/$grp.mod $tmp
pushd $tmp > /dev/null
echo source: $ssSrc.ss
$PHASTBIN/phyloP --method LRT --mode CONACC --wig-scores --chrom $cName \
    -i SS $useGrp $ssSrc.ss > $ssFile.wigFix
popd > /dev/null
mkdir -p $out:h
sleep 4
mv $tmp/$ssFile.wigFix $out
rm -fr $tmp
'_EOF_'
    # << happy emacs
    chmod +x doPhyloP.csh

    # Create list of chunks
    find ../ss -type f | sed -e "s/.ss$//; s#../ss/##;" > ss.list

    # Create template file
    #	file1 == $chr/$chunk/file name without .ss suffix
    cat << '_EOF_' > template
#LOOP
../run.phyloP/doPhyloP.csh $(path1) {check out line+ wigFix/$(dir1)/$(file1).wigFix}
#ENDLOOP
'_EOF_'
    # << happy emacs

    ######################   Running all species  #######################
    # setup run for all species
    mkdir /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/all
    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/all
    rm -fr wigFix
    mkdir wigFix

    gensub2 ../run.phyloP/ss.list single ../run.phyloP/template jobList
    para create jobList
    para try ... check ... push ... etc ...
    para time
# Completed: 2708 of 2708 jobs
# CPU time in finished jobs:    1832980s   30549.67m   509.16h   21.22d  0.058 y
# IO & Wait Time:                217434s    3623.90m    60.40h    2.52d  0.007 y
# Average job time:                 757s      12.62m     0.21h    0.01d
# Longest finished job:            1458s      24.30m     0.41h    0.02d
# Submission to last job:          3647s      60.78m     1.01h    0.04d

    # missed chrM in the original run:
    ../run.phyloP/doPhyloP.csh chrM/chrM.1-16296 wigFix/chrM/chrM.1-16296.wigFix

    ssh hgwdev
    cd /cluster/data/mm10/bed/multiz60way/consPhyloP/run.phyloP/all
    mkdir downloads
    for D in `ls -d wigFix/chr* | sed -e 's#wigFix/##'`
do
    echo "working: $D"
    find ./wigFix/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phyloP60way.wigFix.gz
done
    #   real    38m15.538s

    zcat downloads/*.wigFix.gz \
	| wigEncode stdin phyloP60way.wig phyloP60way.wib > wigEncode.log 2>&1 &
    #   Converted stdin, upper limit 7.53, lower limit -20.00
    #   real    27m53.384s

export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG
    time (zcat downloads/*.wigFix.gz \
        | wigToBigWig stdin ../../../../chrom.sizes phyloP60way.bw)
    #   real    30m10.440s

    bigWigInfo phyloP60way.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 4,533,501,426
primaryIndexSize: 100,775,272
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.169761
min: -20.000000
max: 7.532000
std: 0.942744

    #	if you wanted to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phyloP60way.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phyloP60wayAll; \
            create table phyloP60wayAll \
		(fileName varchar(255) not null); \
            insert into phyloP60wayAll values
	("/gbdb/mm10/bbi/phyloP60way.bw");'

    #	loading the wiggle table:
    ln -s `pwd`/phyloP60way.wib /gbdb/mm10/multiz60way
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phyloP60wayAll phyloP60way.wig
    #   real    1m16.934s

    wigTableStats.sh mm10 phyloP60wayAll
# db.table      min max mean count sumData
# mm10.phyloP60wayAll     -20 7.532 0.169761 1929686275 3.27586e+08
#	stdDev viewLimits
#       0.942744 viewLimits=-4.54396:4.88348
    #	that range is: 4.54396+4.88348 = 9.42744 for -hBinSize=0.0942744 below
    #   to get 1,000 bins

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.0942744 -hBinCount=1000 -hMinVal=-4.54396 -verbose=2 \
	    -db=mm10 phyloP60wayAll > histogram.data 2>&1
    #   real    real    5m58.309s

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phyloP60way track, all 60 vertebrates"
set xlabel " phyloP60way score, all 60 vertebrates"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.2]
set xrange [-2:2]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ######################   Running the glire  #######################
    mkdir /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/glire
    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/glire
    rm -fr wigFix
    mkdir wigFix

    gensub2 ../run.phyloP/ss.list single ../run.phyloP/template jobList
    para create jobList
    para try ... check ... push ... etc ...
    para time
# Completed: 2709 of 2709 jobs
# CPU time in finished jobs:     206723s    3445.39m    57.42h    2.39d  0.007 y
# IO & Wait Time:                256366s    4272.76m    71.21h    2.97d  0.008 y
# Average job time:                 171s       2.85m     0.05h    0.00d
# Longest finished job:             487s       8.12m     0.14h    0.01d
# Submission to last job:          1926s      32.10m     0.54h    0.02d

    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/glire
    mkdir downloads
    for D in `ls -d wigFix/chr* | sed -e 's#wigFix/##'`
do
    echo "working: $D"
    find ./wigFix/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.glire.wigFix.gz
done

    time (zcat downloads/chr*.wigFix.gz  \
	| wigEncode stdin phyloP60wayGlire.wig phyloP60wayGlire.wib \
	> wigEncode.log 2>&1) &
    #   Converted stdin, upper limit 1.17, lower limit -4.35
    #   real     20m31.753s

export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG
    time (zcat downloads/chr*.wigFix.gz \
	| wigToBigWig stdin ../../../../chrom.sizes phyloP60wayGlire.bw) &
    #   real    37m9.063s
    bigWigInfo phyloP60wayGlire.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 3,158,091,915
primaryIndexSize: 100,775,272
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.073187
min: -4.346000
max: 1.165000
std: 0.602992

    #	if you wanted to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phyloP60wayGlire.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phyloP60wayGlire; \
            create table phyloP60wayGlire \
		(fileName varchar(255) not null); \
            insert into phyloP60wayGlire values
	("/gbdb/mm10/bbi/phyloP60wayGlire.bw");'

    #	loading the wiggle table:
    ln -s `pwd`/phyloP60wayGlire.wib /gbdb/mm10/multiz60way
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phyloP60wayGlire phyloP60wayGlire.wig
    #   real    0m58.536s

    wigTableStats.sh mm10 phyloP60wayGlire
# db.table      min max mean count
# mm10.phyloP60wayGlire -4.346 1.165 0.0731873 1929686275 1.41229e+08
#	stdDev viewLimits
#       0.602992 viewLimits=-2.94177:1.165
    #	that range is: 4.346+1.165 = 5.511 -> hBinSize=0.005511

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.005511 -hBinCount=1000 -hMinVal=-4.346 -verbose=2 \
	    -db=mm10 phyloP60wayGlire > histogram.data 2>&1
    #   real    8m23.088s

    #	create plot of histogram:
    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Human Mm10 Histogram phyloP60wayGlire track"
set xlabel " phyloP60wayGlire score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.15]
set xrange [-2:1.2]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ###################   Running the euarchontoglires  #######################
    mkdir /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/euarchontoglires
    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/euarchontoglires
    rm -fr wigFix
    mkdir wigFix

    gensub2 ../run.phyloP/ss.list single ../run.phyloP/template jobList
    para create jobList
    para try ... check ... push ... etc ...
    para time
# Completed: 2709 of 2709 jobs
# CPU time in finished jobs:     542547s    9042.45m   150.71h    6.28d  0.017 y
# IO & Wait Time:                 75914s    1265.23m    21.09h    0.88d  0.002 y
# Average job time:                 228s       3.80m     0.06h    0.00d
# Longest finished job:             430s       7.17m     0.12h    0.00d
# Submission to last job:          4149s      69.15m     1.15h    0.05d

    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/euarchontoglires
    mkdir downloads
    for D in `ls -d wigFix/chr* | sed -e 's#wigFix/##'`
do
    echo "working: $D"
    find ./wigFix/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.euarchontoglires.wigFix.gz
done

    time (zcat downloads/chr*.wigFix.gz  \
	| wigEncode stdin phyloP60wayEuarchontoGlires.wig phyloP60wayEuarchontoGlires.wib \
	> wigEncode.log 2>&1) &
    #   Converted stdin, upper limit 1.75, lower limit -12.70
    #   real    10m52.064s

export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG
    time (zcat downloads/chr*.wigFix.gz \
	| wigToBigWig stdin ../../../../chrom.sizes phyloP60wayEuarchontoGlires.bw) &
    #   real    26m47.912s
    bigWigInfo phyloP60wayEuarchontoGlires.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 3,970,501,521
primaryIndexSize: 100,775,272
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.078739
min: -12.704000
max: 1.753000
std: 0.689759

    #	if you wanted to use the bigWig file, loading bigWig table:
    ln -s `pwd`/phyloP60wayEuarchontoGlires.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phyloP60wayEuarchontoGlires; \
            create table phyloP60wayEuarchontoGlires \
		(fileName varchar(255) not null); \
            insert into phyloP60wayEuarchontoGlires values
	("/gbdb/mm10/bbi/phyloP60wayEuarchontoGlires.bw");'

    #	loading the wiggle table:
    ln -s `pwd`/phyloP60wayEuarchontoGlires.wib /gbdb/mm10/multiz60way
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phyloP60wayEuarchontoGlires phyloP60wayEuarchontoGlires.wig
    #   real    0m51.777s

    time wigTableStats.sh mm10 phyloP60wayEuarchontoGlires
# db.table      min max mean count
# mm10.phyloP60wayEuarchontoGlires -12.704 1.753 0.0787387 1929686275
#	sumData stdDev viewLimits
#       1.51941e+08 0.689759 viewLimits=-3.37006:1.753
    #   real    0m26.197s

    #	that range is: 12.704+1.753 = 14.457 -> hBinSize=0.014457

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.014457 -hBinCount=1000 -hMinVal=-12.704 -verbose=2 \
	    -db=mm10 phyloP60wayEuarchontoGlires > histogram.data 2>&1
    #   real    3m22.205s

    #	create plot of histogram:
    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phyloP60wayEuarchontoGlires track"
set xlabel " phyloP60wayEuarchontoGlires score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.15]
set xrange [-2:1.2]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ######################   Running the primate  #######################
    ### ***### This was constructed
    ### and examined, but not used in the release
    mkdir /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/primate
    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/primate
    rm -fr wigFix
    mkdir wigFix

    gensub2 ../run.phyloP/ss.list single ../run.phyloP/template jobList
    para -ram=8g create jobList
    para try ... check ... push ... etc ...
    para time
# Completed: 2709 of 2709 jobs
# CPU time in finished jobs:     307901s    5131.68m    85.53h    3.56d  0.010 y
# IO & Wait Time:                 42937s     715.62m    11.93h    0.50d  0.001 y
# Average job time:                 130s       2.16m     0.04h    0.00d
# Longest finished job:             234s       3.90m     0.07h    0.00d
# Submission to last job:          5975s      99.58m     1.66h    0.07d

    cd /cluster/data/mm10/bed/multiz60way/consPhyloP/run.phyloP/primate
    mkdir downloads
    for D in `ls -d wigFix/chr* | sed -e 's#wigFix/##'`
do
    echo "working: $D"
    find ./wigFix/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.primate.wigFix.gz
done

    time (zcat downloads/chr*.wigFix.gz \
	| wigEncode stdin phyloP60wayPrimate.wig phyloP60wayPrimate.wib \
	> wigEncode.log 2>&1) &
    #   real    9m37.055s
    #   Converted stdin, upper limit 0.93, lower limit -10.63
export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG
    time (zcat downloads/chr*.wigFix.gz \
	| wigToBigWig stdin ../../../../chrom.sizes phyloP60wayPrimate.bw) &
    #   real    24m18.842s
    bigWigInfo phyloP60wayPrimate.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 2,715,332,211
primaryIndexSize: 100,775,272
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.060017
min: -10.633000
max: 0.930000
std: 0.518027

    #	loading bigWig table:
    ln -s `pwd`/phyloP60wayPrimate.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phyloP60wayPrimate; \
            create table phyloP60wayPrimate \
		(fileName varchar(255) not null); \
            insert into phyloP60wayPrimate values
	("/gbdb/mm10/bbi/phyloP60wayPrimate.bw");'

    #	loading the wiggle table:
    ln -s `pwd`/phyloP60wayPrimate.wib /gbdb/mm10/multiz60way
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phyloP60wayPrimate phyloP60wayPrimate.wig
    #   real    0m45.837s

    wigTableStats.sh mm10 phyloP60wayPrimate
# db.table      min max mean count sumData stdDev viewLimits
# mm10.phyloP60wayPrimate  -10.633 0.93 0.0600168 1929686275 1.15814e+08
#	stdDev viewLimits
#       0.518027 viewLimits=-2.53012:0.93
    #	that range is: 10.633+0.93 = 11.563 for the hBinSize=0.11563

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.11563 -hBinCount=1000 -hMinVal=-10.633 -verbose=2 \
	    -db=mm10 phyloP60wayPrimate > histogram.data 2>&1
    #   real    4m36.379s
    # to see yrange:
    grep -v "^#" histogram.data | ave -col=5 stdin

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Human Mm10 Histogram phyloP60wayPrimate track"
set xlabel " phyloP60wayPrimate score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.472]
set xrange [-2.5:1.0]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ######################   Running the placental  #######################
    mkdir /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/placental
    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/placental
    rm -fr wigFix
    mkdir wigFix

    gensub2 ../run.phyloP/ss.list single ../run.phyloP/template jobList
    para create jobList
    para try ... check ... push ... etc ...
    para time
# Completed: 2709 of 2709 jobs
# CPU time in finished jobs:    1188036s   19800.60m   330.01h   13.75d  0.038 y
# IO & Wait Time:                209859s    3497.65m    58.29h    2.43d  0.007 y
# Average job time:                 516s       8.60m     0.14h    0.01d
# Longest finished job:            1672s      27.87m     0.46h    0.02d
# Submission to last job:          6336s     105.60m     1.76h    0.07d

    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/placental
    mkdir downloads
    for D in `ls -d wigFix/chr* | sed -e 's#wigFix/##'`
do
    echo "working: $D"
    find ./wigFix/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.placental.wigFix.gz
done
    time (zcat downloads/chr*.wigFix.gz \
	| wigEncode stdin phyloP60wayPlacental.wig phyloP60wayPlacental.wib \
	> wigEncode.log 2>&1) &
    #	Converted stdin, upper limit 3.30, lower limit -20.00
    #   real    11m54.289s

export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG
    time (zcat downloads/chr*.wigFix.gz \
	| wigToBigWig stdin ../../../../chrom.sizes phyloP60wayPlacental.bw \           > bigWig.log 2>&1) &
    #   real    28m4.576s
    bigWigInfo phyloP60wayPlacental.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 4,423,832,009
primaryIndexSize: 100,775,272
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.109489
min: -20.000000
max: 3.296000
std: 0.810657

    #	loading bigWig table if that is what you wanted to do:
    ln -s `pwd`/phyloP60wayPlacental.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phyloP60wayPlacental; \
            create table phyloP60wayPlacental \
		(fileName varchar(255) not null); \
            insert into phyloP60wayPlacental values
	("/gbdb/mm10/bbi/phyloP60wayPlacental.bw");'

    #	loading the wiggle table:
    ln -s `pwd`/phyloP60wayPlacental.wib /gbdb/mm10/multiz60way
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phyloP60wayPlacental phyloP60wayPlacental.wig
    #   real    0m50.284s

    wigTableStats.sh mm10 phyloP60wayPlacental
# db.table      min max mean count sumData
# mm10.phyloP60wayPlacental -20 3.296 0.109489 1929686275 2.11279e+08
#       stdDev viewLimits
#       0.810657 viewLimits=-3.9438:3.296

    #	that range is: 20+3.296 = 23.296 for hBinSize=0.023296
    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.023296 -hBinCount=1000 -hMinVal=-20 -verbose=2 \
	    -db=mm10 phyloP60wayPlacental > histogram.data 2>&1
    #   real    3m24.650s
    # to see yrange:
    grep -v "^#" histogram.data | ave -col=5 stdin

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phyloP60wayPlacental track"
set xlabel " phyloP60wayPlacental score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.084]
set xrange [-2.5:2.5]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

    ######################   Running the vertebrate  #######################
    mkdir /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/vertebrate
    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/vertebrate
    rm -fr wigFix
    mkdir wigFix

    gensub2 ../run.phyloP/ss.list single ../run.phyloP/template jobList
    para create jobList
    para try ... check ... push ... etc ...
    para time
# Completed: 2709 of 2709 jobs
# CPU time in finished jobs:    1825414s   30423.56m   507.06h   21.13d  0.058 y
# IO & Wait Time:                211040s    3517.34m    58.62h    2.44d  0.007 y
# Average job time:                 752s      12.53m     0.21h    0.01d
# Longest finished job:            1530s      25.50m     0.42h    0.02d
# Submission to last job:          6045s     100.75m     1.68h    0.07d

    cd /hive/data/genomes/mm10/bed/multiz60way/consPhyloP/vertebrate
    mkdir downloads
    for D in `ls -d wigFix/chr* | sed -e 's#wigFix/##'`
do
    echo "working: $D"
    find ./wigFix/${D} -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \
	| sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \
        | gzip -c > downloads/${D}.phastCons60way.vertebrate.wigFix.gz
done

    time (zcat downloads/chr*.wigFix.gz \
	| wigEncode stdin phyloP60wayVertebrate.wig phyloP60wayVertebrate.wib \
	> wigEncode.log 2>&1) &
    #	Converted stdin, upper limit 7.53, lower limit -20.00
    #   real    12m2.774s

export sizeG=188743680
ulimit -d $sizeG
ulimit -v $sizeG
    time (zcat downloads/chr*.wigFix.gz \
	| wigToBigWig stdin ../../../../chrom.sizes phyloP60wayVertebrate.bw \
        > bigWig.log 2>&1) &
    #   real    27m6.791s
    bigWigInfo phyloP60wayVertebrate.bw
version: 4
isCompressed: yes
isSwapped: 0
primaryDataSize: 4,529,467,614
primaryIndexSize: 100,775,272
zoomLevels: 10
chromCount: 59
basesCovered: 1,929,686,275
mean: 0.169653
min: -20.000000
max: 7.532000
std: 0.942808

    #	loading bigWig table:
    ln -s `pwd`/phyloP60wayVertebrate.bw /gbdb/mm10/bbi
    hgsql mm10 -e 'drop table if exists phyloP60wayVertebrate; \
            create table phyloP60wayVertebrate \
		(fileName varchar(255) not null); \
            insert into phyloP60wayVertebrate values
	("/gbdb/mm10/bbi/phyloP60wayVertebrate.bw");'

    #	loading the wiggle table:
    ln -s `pwd`/phyloP60wayVertebrate.wib /gbdb/mm10/multiz60way
    time nice -n +19 hgLoadWiggle -pathPrefix=/gbdb/mm10/multiz60way mm10 \
	phyloP60wayVertebrate phyloP60wayVertebrate.wig
    #   real    0m56.535s

    time wigTableStats.sh mm10 phyloP60wayVertebrate
# db.table      min max mean count sumData stdDev viewLimits
# mm10.phyloP60wayVertebrate -20 7.532 0.169653 1929686275 3.27377e+08
#	stdDev viewLimits
#       0.942808 viewLimits=-4.54439:4.88369
    #   real    0m25.320s

    #	that range is: 20+7.532 = 27.532 for hBinSize=0.027532

    #  Create histogram to get an overview of all the data
    time nice -n +19 hgWiggle -doHistogram \
	-hBinSize=0.027532 -hBinCount=1000 -hMinVal=-20 -verbose=2 \
	    -db=mm10 phyloP60wayVertebrate > histogram.data 2>&1
    #   real    3m26.565s
    # to see yrange:
    grep -v "^#" histogram.data | ave -col=5 stdin

    #	create plot of histogram:

    cat << '_EOF_' | gnuplot > histo.png
set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff
set size 1.4, 0.8
set key left box
set grid noxtics
set grid ytics
set title " Mouse Mm10 Histogram phyloP60wayVertebrate track"
set xlabel " phyloP60wayVertebrate score"
set ylabel " Relative Frequency"
set y2label " Cumulative Relative Frequency (CRF)"
set y2range [0:1]
set y2tics
set yrange [0:0.0913]
set xrange [-2.5:2.5]

plot "histogram.data" using 2:5 title " RelFreq" with impulses, \
        "histogram.data" using 2:7 axes x1y2 title " CRF" with lines
'_EOF_'
    #	<< happy emacs

    display histo.png &

#########################################################################
# construct download files for 60-way (DONE - 2012-06-27 - 2012-08-21 - Hiram)
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/multiz60way
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/multiz60way/maf
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/multiz60way/alignments
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/glire
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/primate
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/euarchontoglire
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/placental
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/vertebrate
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/mm10.60way.phastCons
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/glire
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/primate
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/euarchontoglire
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/placental
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/vertebrate
mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/mm10.60way.phyloP60way
    mkdir /hive/data/genomes/mm10/bed/multiz60way/downloads
    cd /hive/data/genomes/mm10/bed/multiz60way/downloads
    mkdir multiz60way phastCons60way phyloP60way
    cd multiz60way
    mkdir maf alignments
    cd maf
    time cp -p ../../../anno/result/chr*.maf .
    #   real    735m35.723s
    time gzip *.maf
    #   real    700m23.340s
    md5sum *.maf.gz > md5sum.txt
    ln -s `pwd`/*.maf.gz `pwd`/md5sum.txt \
        /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/multiz60way/maf
    cd ..
    du -hsc maf
    #   24G     maf
    du -hsc ../../anno/result/
    #   244G    ../../anno/result/
    ln -s ../../mm10.60way.nh .
    ln -s ../../mm10.60way.commonNames.nh .
    ln -s `pwd`/*.nh \
        /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/multiz60way

    #####################################################################
    cd /hive/data/genomes/mm10/bed/multiz60way/downloads/phastCons60way
    mkdir glire euarchontoglire primate placental vertebrate mm10.60way.phastCons
    cd glire
    ln -s ../../../cons/glire/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
        /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/glire
    #   real    5m50.001s
    cd ../euarchontoglire
    ln -s ../../../cons/euarchontoglires/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    #   real    1m14.103s
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
      /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/euarchontoglire
    cd ../primate
    ln -s ../../../cons/primate/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
      /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/primate
    #   real    5m39.288s
    cd ../placental
    ln -s ../../../cons/placental/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
  /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/placental
    #   real    5m9.762s
    cd ../vertebrate
    ln -s ../../../cons/vertebrate/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
  /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/vertebrate
    #   real    0m45.408s
    cd ../mm10.60way.phastCons
    ln -s ../../../cons/all/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
  /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way/mm10.60way.phastCons
    #   real    6m11.158s
    cd ..
    ln -s ../../cons/all/all.mod mm10.60way.phastCons.mod
    ln -s ../../cons/glire/glire.mod mm10.60way.phastCons.glire.mod
    ln -s ../../cons/primate/primate.mod mm10.60way.phastCons.primate.mod
    ln -s ../../cons/euarchontoglires/euarchontoglires.mod mm10.60way.phastCons.euarchontoglire.mod
    ln -s ../../cons/placental/placental.mod mm10.60way.phastCons.placental.mod
    ln -s ../../cons/vertebrate/vertebrate.mod mm10.60way.phastCons.vertebrate.mod
    ln -s ../../cons/all/phastCons60way.bw mm10.60way.phastCons.bw
    ln -s ../../cons/glire/phastCons60wayGlire.bw \
        mm10.60way.phastCons60wayGlire.bw
    ln -s ../../cons/placental/phastCons60wayPlacental.bw \
        mm10.60way.phastCons60wayPlacental.bw
    ln -s ../../cons/euarchontoglires/phastCons60wayEuarchontoGlires.bw \
        mm10.60way.phastCons60wayEuarchontoGlire.bw
    ln -s ../../cons/primate/phastCons60wayPrimate.bw \
        mm10.60way.phastCons60wayPrimate.bw
    ln -s ../../cons/vertebrate/phastCons60wayVertebrate.bw \
        mm10.60way.phastCons60wayVertebrate.bw
    time md5sum *.mod *.bw > md5sum.txt
    #   real    20m11.260s
    # obtain the README.txt from hg19/phastCons46way and update for this
    #   situation
    ln -s `pwd`/*.mod `pwd`/*.bw `pwd`/README.txt \
      /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phastCons60way

    #####################################################################
    cd /hive/data/genomes/mm10/bed/multiz60way/downloads/phyloP60way
    mkdir glire euarchontoglire primate placental vertebrate mm10.60way.phyloP60way
    cd glire
    ln -s ../../../consPhyloP/glire/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
        /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/glire
    #   real    6m5.733s
    cd ../euarchontoglire
    ln -s ../../../consPhyloP/euarchontoglires/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
/usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/euarchontoglire
    #   real    5m40.272s
    cd ../primate
    ln -s ../../../consPhyloP/primate/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
        /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/primate
    #   real    7m22.623s
    cd ../placental
    ln -s ../../../consPhyloP/placental/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
      /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/placental
    #   real    7m39.269s
    cd ../vertebrate
    ln -s ../../../consPhyloP/vertebrate/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
      /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/vertebrate
    cd ../mm10.60way.phyloP60way
    ln -s ../../../consPhyloP/all/downloads/chr*.gz .
    time md5sum *.gz > md5sum.txt &
    ln -s `pwd`/*.gz `pwd`/md5sum.txt \
      /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way/mm10.60way.phyloP60way
    #   real    8m5.777s

    cd ..
    ln -s ../../consPhyloP/run.phyloP/all.mod mm10.60way.phyloP60way.mod
    ln -s ../../consPhyloP/run.phyloP/glire.mod ./mm10.phyloP.glire.mod
    ln -s ../../consPhyloP/run.phyloP/placental.mod ./mm10.phyloP.placental.mod
    ln -s ../../consPhyloP/run.phyloP/euarchontoglires.mod ./mm10.phyloP.euarchontoglire.mod
    ln -s ../../consPhyloP/run.phyloP/primate.mod ./mm10.phyloP.primate.mod
    ln -s ../../consPhyloP/run.phyloP/vertebrate.mod ./mm10.60way.vertebrate.mod

    ln -s ../../consPhyloP/all/phyloP60way.bw mm10.60way.phyloP60way.bw
    ln -s ../../consPhyloP/glire/phyloP60wayGlire.bw \
        mm10.60way.phyloP60wayGlire.bw
    ln -s ../../consPhyloP/vertebrate/phyloP60wayVertebrate.bw \
        mm10.60way.phyloP60wayVertebrate.bw
    ln -s ../../consPhyloP/placental/phyloP60wayPlacental.bw \
        mm10.60way.phyloP60wayPlacental.bw
    ln -s ../../consPhyloP/euarchontoglires/phyloP60wayEuarchontoGlires.bw \
        mm10.60way.phyloP60wayEuarchontoglire.bw
    ln -s ../../consPhyloP/primate/phyloP60wayPrimate.bw \
        mm10.60way.phyloP60wayPrimate.bw

    time md5sum *.mod *.bw > md5sum.txt &
    #   real    20m17.082s

    # obtain the README.txt from hg19/phyloP46way and update for this
    #   situation
    ln -s `pwd`/*.mod `pwd`/*.bw `pwd`/md5sum.txt `pwd`/README.txt \
      /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/phyloP60way

    ###########################################################################
    ## create upstream refGene maf files
    cd /hive/data/genomes/mm10/bed/multiz60way/downloads/maf
    # bash script
#!/bin/sh
for S in 1000 2000 5000
do
    echo "making upstream${S}.maf"
    featureBits mm10 refGene:upstream:${S} -fa=/dev/null -bed=stdout \
        | perl -wpe 's/_up[^\t]+/\t0/' | sort -k1,1 -k2,2n \
        | /cluster/bin/$MACHTYPE/mafFrags mm10 multiz60way \
                stdin stdout \
                -orgs=/hive/data/genomes/mm10/bed/multiz60way/species.list \
        | gzip -c > upstream${S}.maf.gz
    echo "done upstream${S}.maf.gz"
done
    #   real    199m45.558s

    md5sum *.nh *.maf.gz > md5sum.txt
    #   real    27m59.778s

    # obtain the README.txt from hg19/multiz46way and update for this
    #   situation
    ln -s `pwd`/*.nh `pwd`/*.maf.gz `pwd`/*.txt \
        /usr/local/apache/htdocs-hgdownload/goldenPath/mm10/multiz60way

#############################################################################
# hgPal downloads (DONE - 2012-07-05 - 2012-07-09 - Hiram)
#   FASTA from 60-way for refGene

    ssh hgwdev
    screen -S mm10HgPal
    mkdir /hive/data/genomes/mm10/bed/multiz60way/pal
    cd /hive/data/genomes/mm10/bed/multiz60way/pal
    cat ../species.list | tr '[ ]' '[\n]' > order.list

    export mz=multiz60way
    export gp=refGene
    export db=mm10
    export I=0
    mkdir exonAA exonNuc
    for C in `sort -nk2 ../../../chrom.sizes | cut -f1`
    do
        I=`echo $I | awk '{print $1+1}'`
	echo "mafGene -chrom=$C -exons -noTrans $db $mz $gp order.list stdout | gzip -c > exonNuc/$C.exonNuc.fa.gz &"
	echo "mafGene -chrom=$C -exons $db $mz $gp order.list stdout | gzip -c > exonAA/$C.exonAA.fa.gz &"
        if [ $I -gt 6 ]; then
            echo "date"
            echo "wait"
            I=0
        fi
    done > $gp.jobs
    echo "date" >> $gp.jobs
    echo "wait" >> $gp.jobs

    time sh -x $gp.jobs > $gp.jobs.log 2>&1 &
    #   real    93m34.376s

    mz=multiz60way
    gp=refGene
    db=mm10
    time zcat exonAA/*.gz | gzip -c > $gp.$mz.exonAA.fa.gz
    #   real    1m16.821s
    zcat exonNuc/*.gz | gzip -c > $gp.$mz.exonNuc.fa.gz

    rm -rf exonAA exonNuc

    # we're only distributing exons at the moment
    mz=multiz60way
    gp=refGene
    db=mm10
    pd=/usr/local/apache/htdocs-hgdownload/goldenPath/$db/$mz/alignments
    mkdir -p $pd
    md5sum *.fa.gz > md5sum.txt
    ln -s `pwd`/$gp.$mz.exonAA.fa.gz $pd/$gp.exonAA.fa.gz
    ln -s `pwd`/$gp.$mz.exonNuc.fa.gz $pd/$gp.exonNuc.fa.gz
    ln -s `pwd`/md5sum.txt $pd/


#########################################################################
# lastz nile tilapia oreNil2 (DONE - 2012-04-02 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10OreNil2
    mkdir /hive/data/genomes/mm10/bed/lastzOreNil2.2012-04-11
    cd /hive/data/genomes/mm10/bed/lastzOreNil2.2012-04-11

    cat << '_EOF_' > DEF
# Mouse vs. nile tilapia
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: nile tilapia oreNil2
SEQ2_DIR=/hive/data/genomes/oreNil2/oreNil2.2bit
SEQ2_LEN=/hive/data/genomes/oreNil2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=10

BASE=/hive/data/genomes/mm10/bed/lastzOreNil2.2012-04-11
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #   real    108m51.232s

    cat fb.mm10.chainOreNil2Link.txt
    #   51909908 bases of 2652783500 (1.957%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzOreNil2.2012-04-11 lastz.oreNil2

    #	and for the swap
    mkdir /hive/data/genomes/oreNil2/bed/blastz.mm10.swap
    cd /hive/data/genomes/oreNil2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzOreNil2.2012-04-11/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #   real    9m8.213s
    cat  fb.oreNil2.chainMm10Link.txt
    #   49704887 bases of 816084674 (6.091%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/oreNil2/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# LASTZ pig susScr3 (DONE - 2012-04-13 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10SusScr3
    mkdir /hive/data/genomes/mm10/bed/lastzSusScr3.2012-04-13
    cd /hive/data/genomes/mm10/bed/lastzSusScr3.2012-04-13

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# pig vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: pig SusScr3
SEQ2_DIR=/hive/data/genomes/susScr3/susScr3.2bit
SEQ2_LEN=/hive/data/genomes/susScr3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzSusScr3.2012-04-13
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    1086m29.992s
    cat fb.mm10.chainSusScr3Link.txt
    #   681359766 bases of 2652783500 (25.685%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzSusScr3.2012-04-13 lastz.susScr3

    mkdir /hive/data/genomes/susScr3/bed/blastz.mm10.swap
    cd /hive/data/genomes/susScr3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzSusScr3.2012-04-13/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #   real    104m56.258s
    cat fb.susScr3.chainMm10Link.txt
    #   743574150 bases of 2525294057 (29.445%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/susScr3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ armadillo dasNov3 (DONE - 2012-04-13 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10DasNov3
    mkdir /hive/data/genomes/mm10/bed/lastzDasNov3.2012-04-13
    cd /hive/data/genomes/mm10/bed/lastzDasNov3.2012-04-13

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# armadillo vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: armadillo DasNov3
SEQ2_DIR=/hive/data/genomes/dasNov3/dasNov3.2bit
SEQ2_LEN=/hive/data/genomes/dasNov3/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=200

BASE=/hive/data/genomes/mm10/bed/lastzDasNov3.2012-04-13
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    1125m34.124s
    cat fb.mm10.chainDasNov3Link.txt
    #   668529920 bases of 2652783500 (25.201%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzDasNov3.2012-04-13 lastz.dasNov3

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzDasNov3.2012-04-13
    time doRecipBest.pl mm10 dasNov3 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #   real    116m51.114s

    mkdir /hive/data/genomes/dasNov3/bed/blastz.mm10.swap
    cd /hive/data/genomes/dasNov3/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzDasNov3.2012-04-13/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	real    150m51.653s
    cat fb.dasNov3.chainMm10Link.txt
    #   695161920 bases of 3299882059 (21.066%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/dasNov3/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ cat felCat5 (DONE - 2012-04-13 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10FelCat5
    mkdir /hive/data/genomes/mm10/bed/lastzFelCat5.2012-04-13
    cd /hive/data/genomes/mm10/bed/lastzFelCat5.2012-04-13

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# cat vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: cat FelCat5
SEQ2_DIR=/hive/data/genomes/felCat5/felCat5.2bit
SEQ2_LEN=/hive/data/genomes/felCat5/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzFelCat5.2012-04-13
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    1029m54.494s
    cat fb.mm10.chainFelCat5Link.txt
    #   788544084 bases of 2652783500 (29.725%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzFelCat5.2012-04-13 lastz.felCat5

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzFelCat5.2012-04-13
    time doRecipBest.pl mm10 felCat5 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #	real    106m30.011s

    mkdir /hive/data/genomes/felCat5/bed/blastz.mm10.swap
    cd /hive/data/genomes/felCat5/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzFelCat5.2012-04-13/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #	 real 124m25.850s
    cat fb.felCat5.chainMm10Link.txt
    #   762344436 bases of 2364296207 (32.244%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/felCat5/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ naked mole rat hetGla2 (DONE - 2012-04-14 - Hiram)
    #	establish a screen to control this job
    screen -S mm10HetGla2
    mkdir /hive/data/genomes/mm10/bed/lastzHetGla2.2012-04-14
    cd /hive/data/genomes/mm10/bed/lastzHetGla2.2012-04-14

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# naked mole rat vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: naked mole rat HetGla2
SEQ2_DIR=/hive/data/genomes/hetGla2/hetGla2.2bit
SEQ2_LEN=/hive/data/genomes/hetGla2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=30

BASE=/hive/data/genomes/mm10/bed/lastzHetGla2.2012-04-14
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    690m7.626s
    cat fb.mm10.chainHetGla2Link.txt
    #	853221843 bases of 2652783500 (32.163%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzHetGla2.2012-04-14 lastz.hetGla2

    mkdir /hive/data/genomes/hetGla2/bed/blastz.mm10.swap
    cd /hive/data/genomes/hetGla2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzHetGla2.2012-04-14/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #   real    92m24.775s
    cat fb.hetGla2.chainMm10Link.txt
    #   879356778 bases of 2314771103 (37.989%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/hetGla2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ dolphin turTru2 (DONE - 2012-04-14 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10TurTru2
    mkdir /hive/data/genomes/mm10/bed/lastzTurTru2.2012-04-14
    cd /hive/data/genomes/mm10/bed/lastzTurTru2.2012-04-14

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# dolphin vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: dolphin TurTru2
SEQ2_DIR=/hive/data/genomes/turTru2/turTru2.2bit
SEQ2_LEN=/hive/data/genomes/turTru2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=500

BASE=/hive/data/genomes/mm10/bed/lastzTurTru2.2012-04-14
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    624m36.508s
    cat fb.mm10.chainTurTru2Link.txt
    #   802921354 bases of 2652783500 (30.267%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzTurTru2.2012-04-14 lastz.turTru2

    # better to have reciprocal best for this one since it is low coverage:
    cd /hive/data/genomes/mm10/bed/lastzTurTru2.2012-04-14
    time doRecipBest.pl mm10 turTru2 -buildDir=`pwd` -workhorse=hgwdev \
	> best.log 2>&1 &
    #   real    44m47.753s

    mkdir /hive/data/genomes/turTru2/bed/blastz.mm10.swap
    cd /hive/data/genomes/turTru2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzTurTru2.2012-04-14/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #   real    124m17.088s
    cat fb.turTru2.chainMm10Link.txt
    #   781169007 bases of 2332402443 (33.492%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/turTru2/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# LASTZ Gibbon nomLeu2 (DONE - 2012-04-14 - Hiram)
    screen -S mm10NomLeu2
    mkdir /hive/data/genomes/mm10/bed/lastzNomLeu2.2012-04-14
    cd /hive/data/genomes/mm10/bed/lastzNomLeu2.2012-04-14

    cat << '_EOF_' > DEF
# gibbon vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Gibbon NomLeu2
SEQ2_DIR=/hive/data/genomes/nomLeu2/nomLeu2.2bit
SEQ2_LEN=/hive/data/genomes/nomLeu2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=100

BASE=/hive/data/genomes/mm10/bed/lastzNomLeu2.2012-04-14
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    #	establish a screen to control this job
    screen -S mm10NomLeu2
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	`pwd`/DEF \
	-syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    621m38.251s
    cat fb.mm10.chainNomLeu2Link.txt
    #   902774780 bases of 2652783500 (34.031%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzNomLeu2.2012-04-14 lastz.nomLeu2

    mkdir /hive/data/genomes/nomLeu2/bed/blastz.mm10.swap
    cd /hive/data/genomes/nomLeu2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzNomLeu2.2012-04-14/DEF \
	-swap -syntenicNet \
	-workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #   real 92m24.775s
    cat fb.nomLeu2.chainMm10Link.txt
    #   889660339 bases of 2756609047 (32.274%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/nomLeu2/bed
    ln -s blastz.mm10.swap lastz.mm10

#####################################################################
# tRNAs track (DONE 2012-04-02 Chin)
#
    # Please refer to the generic tRNS track build documentation
    #   ~/kent/src/hg/makeDb/doc/tRNAsTrack.txt
    # for details about how the track was build.

##############################################################################
# orfeome 2012-03-16  (markd)

enabled ORFeome tracks in etc/genbank.conf and reload genbank

############################################################################
# construct liftOver to mm9 (DONE - 2012-04-30 - Hiram)
    screen -S 10        # manage this longish running job in a screen
    mkdir /hive/data/genomes/mm10/bed/blat.mm9.2012-04-30
    cd /hive/data/genomes/mm10/bed/blat.mm9.2012-04-30
    # check it with -debug first to see if it is going to work:
    time doSameSpeciesLiftOver.pl -buildDir=`pwd` -bigClusterHub=swarm \
        -ooc=/scratch/data/mm10/mm10.11.ooc \
        -debug -dbHost=hgwdev -workhorse=hgwdev mm10 mm9 > do.log 2>&1
    # if that is OK, then run it:
    time doSameSpeciesLiftOver.pl -buildDir=`pwd` -bigClusterHub=swarm \
        -ooc=/scratch/data/mm10/mm10.11.ooc \
        -dbHost=hgwdev -workhorse=hgwdev mm10 mm9 > do.log 2>&1
    #   real    95m21.635s

    # verify this file exists:
    og -L /gbdb/mm10/liftOver/mm10ToMm9.over.chain.gz
# -rw-rw-r-- 1 535855 Feb  9 12:07 /gbdb/mm9/liftOver/mm9ToMm10.over.chain.gz

    # and try out the conversion on genome-test from mm9 to mm10
############################################################################
# EXONIPHY MM10, lifted from hg19 (DONE - braney 2012-05-29)
#	needed for ucscGenes building
    # create a syntenic liftOver chain file
    cd /cluster/data/hg19/bed/lastz.mm10/axtChain
    time nice -n +19 netFilter -syn hg19.mm10.net.gz \
	| netChainSubset -verbose=0 stdin hg19.mm10.all.chain.gz stdout \
	| chainStitchId stdin stdout | gzip -c > hg19.mm10.syn.chain.gz
    #real    2m38.915s
    #user    3m29.458s
    #sys     0m16.033s

    #	slightly smaller than the ordinary liftOver chain file:
-rw-rw-r-- 1 78419424 Mar  7 18:40 hg19.mm10.over.chain.gz
-rw-rw-r-- 1 74588027 May 29 12:29 hg19.mm10.syn.chain.gz

    # exoniphyMm9.gp is prepared as follows
    mkdir /cluster/data/mm10/bed/exoniphy
    cd /cluster/data/mm10/bed/exoniphy
    hgsql hg19 -e "select * from exoniphy" -N | cut  -f 2-16 > exoniphyHg19.gp
    time nice -n +19 liftOver -genePred exoniphyHg19.gp \
	/cluster/data/hg19/bed/lastz.mm10/axtChain/hg19.mm10.syn.chain.gz \
	    exoniphyMm10.gp unmapped
    # real    16m0.334s
    # user    15m46.462s
    # sys     0m7.115s

    wc -l *
    # 186601 exoniphyHg19.gp
    # 178821 exoniphyMm10.gp
    # 15560 unmapped

    cd /cluster/data/mm10/bed/exoniphy
    nice -n +19 hgLoadGenePred -genePredExt mm10 exoniphy exoniphyMm10.gp
    nice -n +19 featureBits mm10 exoniphy
    # 26795543 bases of 2652783500 (1.010%) in intersection
    nice -n +19 featureBits mm9 exoniphy
    #	25931742 bases of 2620346127 (0.990%) in intersection

##############################################################################
# LASTZ cow bosTau6 (DONE - 2012-06-19 - Chin)
    # establish a screen to control this job with a name to indicate
    # what it is
    screen -S mm10BosTau6
    mkdir /hive/data/genomes/mm10/bed/lastzBosTau6.2012-06-19
    cd /hive/data/genomes/mm10/bed/lastzBosTau6.2012-06-19

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #   number of jobs, 50,000 to something under 100,000
    cat << '_EOF_' > DEF
# cow vs mouse
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: cow BosTau6
SEQ2_DIR=/scratch/data/bosTau6/bosTau6.2bit
SEQ2_LEN=/scratch/data/bosTau6/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzBosTau6.2012-06-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
    `pwd`/DEF \
    -syntenicNet \
    -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
    -chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    212m21.604s

    cat fb.mm10.chainBosTau6Link.txt
    #   700039696 bases of 2652783500 (26.389%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzBosTau6.2012-06-19 lastz.bosTau6

    # swap
    mkdir /hive/data/genomes/bosTau6/bed/blastz.mm10.swap
    cd /hive/data/genomes/bosTau6/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
    /hive/data/genomes/mm10/bed/lastzBosTau6.2012-06-19/DEF \
    -swap -syntenicNet \
    -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
    -chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #   real    72m13.925s
    cat fb.bosTau6.chainMm10Link.txt
    #   688651806 bases of 2649682029 (25.990%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/bosTau6/bed
    ln -s blastz.mm10.swap lastz.mm10

##############################################################################
# lastz Medium Ground Finch geoFor1 (DONE - 2012-07-29 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10
    mkdir /hive/data/genomes/mm10/bed/lastzGeoFor1.2012-07-29
    cd /hive/data/genomes/mm10/bed/lastzGeoFor1.2012-07-29

    cat << '_EOF_' > DEF
# Mouse vs. medium ground finch
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Medium Ground Finch GeoFor1
SEQ2_DIR=/hive/data/genomes/geoFor1/geoFor1.2bit
SEQ2_LEN=/hive/data/genomes/geoFor1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=50

BASE=/hive/data/genomes/mm10/bed/lastzGeoFor1.2012-07-29
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #   real    251m4.194s
    cat fb.mm10.chainGeoFor1Link.txt
    #   93984241 bases of 2652783500 (3.543%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzGeoFor1.2012-07-29 lastz.geoFor1

    #	and for the swap
    mkdir /hive/data/genomes/geoFor1/bed/blastz.mm10.swap
    cd /hive/data/genomes/geoFor1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzGeoFor1.2012-07-29/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #   real    10m0.875s
    cat  fb.geoFor1.chainMm10Link.txt
    #   80273915 bases of 1041286029 (7.709%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/geoFor1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# construct assembly fragments table (DONE - 2012-09-11 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/assemblyFrags
    cd /hive/data/genomes/mm10/bed/assemblyFrags
    zgrep -h -v "^#" "${F}"

    zgrep -h -v "^#" ../../genbank/Primary_Assembly/assembled_chromosomes/AGP/*.comp.agp.gz \
    | awk '$5 != "N"' \
        | awk '{printf "%s\t%d\t%d\t%s\t0\t%s\n", $1,$2-1,$3,$6,$9}' \
    | sed -e 's/CM000994.2/chr1/; s/CM000995.2/chr2/; s/CM000996.2/chr3/; s/CM000997.2/chr4/; s/CM000998.2/chr5/; s/CM000999.2/chr6/; s/CM001000.2/chr7/; s/CM001001.2/chr8/; s/CM001002.2/chr9/; s/CM001003.2/chr10/; s/CM001004.2/chr11/; s/CM001005.2/chr12/; s/CM001006.2/chr13/; s/CM001007.2/chr14/; s/CM001008.2/chr15/; s/CM001009.2/chr16/; s/CM001010.2/chr17/; s/CM001011.2/chr18/; s/CM001012.2/chr19/; s/CM001013.2/chrX/; s/CM001014.2/chrY/;' > chr.asmFrag.bed

    zgrep -h -v "^#" ../../genbank/Primary_Assembly/unlocalized_scaffolds/AGP/*.agp.gz \
        | awk '$5 != "N"' \
        | awk '{printf "%s\t%d\t%d\t%s\t0\t%s\n", $1,$2-1,$3,$6,$9}' \
    | sed -e "s#GL456233.1#chrX_GL456233_random#; s#GL456216.1#chr4_GL456216_random#; s#JH584299.1#chr5_JH584299_random#; s#JH584301.1#chrY_JH584301_random#; s#JH584300.1#chrY_JH584300_random#; s#JH584303.1#chrY_JH584303_random#; s#JH584302.1#chrY_JH584302_random#; s#JH584298.1#chr5_JH584298_random#; s#JH584297.1#chr5_JH584297_random#; s#JH584296.1#chr5_JH584296_random#; s#JH584295.1#chr4_JH584295_random#; s#JH584294.1#chr4_JH584294_random#; s#JH584293.1#chr4_JH584293_random#; s#JH584292.1#chr4_JH584292_random#; s#GL456354.1#chr5_GL456354_random#; s#GL456350.1#chr4_GL456350_random#; s#GL456221.1#chr1_GL456221_random#; s#GL456219.1#chr7_GL456219_random#; s#GL456213.1#chr1_GL456213_random#; s#GL456212.1#chr1_GL456212_random#; s#GL456211.1#chr1_GL456211_random#; s#GL456210.1#chr1_GL456210_random#;" > chrUL.asmFrag.bed

    zgrep -h -v "^#" ../../genbank/Primary_Assembly/unplaced_scaffolds/AGP/*.agp.gz \
        | awk '$5 != "N"' | sed -e 's/\.1\t/\t/' \
        | awk '{printf "chrUn_%s\t%d\t%d\t%s\t0\t%s\n", $1,$2-1,$3,$6,$9}' \
        > chrUn.asmFrag.bed

    cat chr.asmFrag.bed chrUL.asmFrag.bed chrUn.asmFrag.bed > mm10.asmFrag.bed
    # add the chrM identity
    echo -e "chrM\t0\t1629\tAY172335.1\t0\t+" >> mm10.asmFrag.bed
    hgLoadBed mm10 assemblyFrags mm10.asmFrag.bed
    featureBits mm10 assemblyFrags
    #   2652769048 bases of 2652783500 (99.999%) in intersection
    # should be silent when all chr names are correct:
    checkTableCoords mm10 assemblyFrags

#########################################################################
# construct ucscToEnsembl table (DONE - 2012-09-11 - Hiram)
    mkdir /hive/data/genomes/mm10/ensembl
    cd /hive/data/genomes/mm10/ensembl
    wget --timestamping \
'ftp://ftp.ensembl.org/pub/release-68/fasta/mus_musculus/dna/Mus_musculus.GRCm38.68.dna.toplevel.fa.gz'

    wget --timestamping \
'ftp://ftp.ensembl.org/pub/release-68/fasta/mus_musculus/dna/Mus_musculus.GRCm38.68.dna.nonchromosomal.fa.gz'

    faCount *.fa.gz > faCount.txt
    egrep -v "total|seq" faCount.txt  | awk '{print $1,$2}' \
        | sort -u | sort -k2nr | sed -e "s/ /\t/" > ensembl.chrom.sizes

    mkdir /hive/data/genomes/mm10/bed/ucscToEnsembl
    cd /hive/data/genomes/mm10/bed/ucscToEnsembl
    awk '{printf "%d\t%s\n", $2,$1}' ../../chrom.sizes | sort > sizes.chrom.ucsc
    awk '{printf "%d\t%s\n", $2,$1}' ../../ensembl/ensembl.chrom.sizes \
        | sort > sizes.chrom.ensembl
    join sizes.chrom.ucsc sizes.chrom.ensembl \
        | awk '{printf "%s\t%s\n", $2,$3}' > ucscToEnsembl.tab

    cut -f1 ucscToEnsembl.tab | awk '{print length($1)}' | sort -rn | head -1
    #   20

    cat << '_EOF_' > ucscToEnsembl.sql
# UCSC to Ensembl chr name translation
CREATE TABLE ucscToEnsembl (
    ucsc varchar(255) not null,        # UCSC chromosome name
    ensembl varchar(255) not null,     # Ensembl chromosome name
              #Indices
    PRIMARY KEY(ucsc(20))
);
'_EOF_'

    hgLoadSqlTab mm10 ucscToEnsembl ucscToEnsembl.sql ucscToEnsembl.tab

#########################################################################
# GRC Incident database (DONE - 2012-09-21 - Hiram)
    # updated the automatic scripts to include the build of this track
    #   on Mm10
    # this procedure is run as a cron job in Hiram's account:

    #	43 09 * * * /hive/data/outside/grc/incidentDb/runUpdate.sh makeItSo

    # using the two scrips there: runUpdate.sh and update.sh
    # which are checked into the source tree as files:
    #	src/hg/utils/automation/grcIncidentUpdate.sh
    #	src/hg/utils/automation/grcRunIncidentUpdate.sh

    # they fetch the XML files from NCBI, convert them to SQL text
    # files, construct a bigBed file, and pushes it to genomewiki if
    # it is an update from previous

    # the table in the dataBase is: grcIncidentDb
    # which is the URL to the bb file, a single row:
    #   http://genomewiki.ucsc.edu/images/a/a4/Mm10.grcIncidentDb.bb

    # construct the table after running the script once manually:
    hgBbiDbLink mm10 grcIncidentDb \
        "http://genomewiki.ucsc.edu/images/a/a4/Mm10.grcIncidentDb.bb"

#########################################################################
# GRCm38.p1 patch 1 (DONE - 2012-09-21 - Hiram)
    mkdir /hive/data/genomes/mm10/bed/patch1
    cd /hive/data/genomes/mm10/bed/patch1
    rsync -a -P rsync://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Mus_musculus/GRCm38.p1/ ./genbank/
    # slight modifications to this script from hg19 patch9 work:
    ./gatherNames.pl genbank > ucscNames.patch1.txt
    # examine the names for sanity:
    awk '{print $NF}' ucscNames.patch1.txt | sort
    # and they should not be longer than 31 characters:
    awk '{print $NF}' ucscNames.patch1.txt | sort | awk '{print length($0)}' \
        | sort -n | tail
    # script from hg19 patch9, update the variable patchName
     ./mkTables.pl patches.chrom.sizes ucscNames.patch1.txt genbank/PATCHES/alt_scaffolds/AGP/alt.scaf.agp.gz
    # output to stdout is the contents of alt.scaf.agp.gz
    # constructs ctgPos.txt chromInfo.txt gap.txt gold.txt
    # script from hg19 patch9, update the variable patchName
    ./mkCtgPos2.pl ucscNames.patch1.txt patches.chrom.sizes > ctgPos2.txt
    cp -p ../patch5/mkHapLocate.pl .
    ./mkHapLocate.pl ctgPos.txt \
	PATCHES/alt_scaffolds/alt_scaffold_placement.txt \
	> haplotypeLocations.bed
    cp -p haplotypeLocations.bed altSequence.bed

    ./mkFasta.pl ucscNames.patch1.txt > mm10.patch1.fa
    # the build of mm10Patch1 can be seen in mm10Patch1.txt

    egrep -v "32,32,190" altSequence.bed  \
	| awk '{printf "%s\t%d\t%d\t%s\t%d\t%s\n", $1,$2,$3,$4,$5,$6}' \
	    > altSeqPatchesP1.tab
    # no haplotypes yet, this is nothing:
    egrep "32,32,190" altSequence.bed  \
	| awk '{printf "%s\t%d\t%d\t%s\t%d\t%s\n", $1,$2,$3,$4,$5,$6}' \
	    > altSeqHaplotypesP1.tab
    # verify none lost
    wc -l altSequence.bed altSeqPatchesP1.tab altSeqHaplotypesP1.tab
    #   9 altSequence.bed
    #   9 altSeqPatchesP1.tab
    #   0 altSeqHaplotypesP1.tab

    # not necessary, there are none yet:
    hgLoadBed mm10 altSeqHaplotypesP1 altSeqHaplotypesP1.tab
    #	Loaded 75 elements of size 6
    hgLoadBed mm10 altSeqPatchesP1 altSeqPatchesP1.tab
    #   Read 9 elements of size 6 from altSeqPatchesP1.tab

    #    these tables are part of mouse/mm10/altSeqComposite1.ra

##############################################################################
# Haplotype track (WORKING - 2012-10-01 - Hiram)

    cat << '_EOF_' > mkBedFile.pl
#!/usr/bin/env perl

use strict;
use warnings;

my $debug = 1;

sub usage() {
    print STDERR "usage: ./mkBedFile.pl ../../mm10/genbank > mm10Haplotypes.bed\n";
    print STDERR "expecting the Mus_musculus/GRCm38.p1/ hierarchy in ./genbank from NCBI\n";
    exit 255;
}

my $argc = scalar(@ARGV);

if ($argc != 1) {
    usage;
}

my $patchDir = shift;

if ( ! -d $patchDir ) {
    print STDERR "ERROR: given directory $patchDir is not a directory or does not exist";
    usage;
}

my %glSize;
my %ctgToChr;
my %ctgToFastaName;
# my $fasta = "$patchDir/PATCHES/alt_scaffolds/FASTA/alt.scaf.fa.gz";
my @placeList = split('\n',`find $patchDir -type f | grep placement.txt | grep alt_scaffolds | grep -v UNKNOWN`);
for (my $i = 0; $i < scalar(@placeList); ++$i) {
    printf STDERR "# %s\n", $placeList[$i];
    open (FH, "grep -v '^#' $placeList[$i]|") or die "can not read $placeList[$i]";
    while (my $line = <FH>) {
#        printf STDERR "%s", $line;
        chomp $line;
        my @a = split('\s+', $line);
        next if ($a[11] eq "na");
        $a[8] = "+" if ($a[8] eq "b");
        my $descr = sprintf("<B>Region&nbsp;name:&nbsp;</B>%s", $a[7]);
        printf "chr%s\t%d\t%d\t%s\t0\t%s\t%s\t%s\n",
                $a[5], $a[11], $a[12], $a[0], $a[8], $a[3], $descr;
    }
    close (FH);
}
'_EOF_'
    # << happy emacs
    chmod +x mkBedFile.pl

    ./mkBedFile.pl > mm10Haplotypes.bedDetail
    cat << '_EOF_' > mm10Haplotypes.sql
CREATE TABLE mm10Haplotypes (
    chrom varchar(255) not null,   # Reference sequence chromosome or scaffold
    chromStart int unsigned not null,   # Start position in chromosome
    chromEnd int unsigned not null,     # End position in chromosome
    name varchar(255) not null, # Short Name of item
    score int unsigned, # Score from 0-1000
    strand char(1),     # + or -
    id varchar(255) not null,   # ID to bed used in URL to link back
    description longblob not null, # Long description of item for the details page
    #Indices
    INDEX(chrom, chromStart)
);
'_EOF_'

    hgLoadSqlTab mm10 mm10Haplotypes mm10Haplotypes.sql mm10Haplotypes.bedDetail

    # trackDb entry:
track mm10Haplotypes
shortLabel Alt. strains
longLabel Alternate mouse strains, mapped to reference as haplotypes
group varRep
priority 111
visibility hide
type bedDetail 8
url http://www.ncbi.nlm.nih.gov/nuccore/$$
urlLabel NCBI Nucleotide:

##########################################################################
##  CYTOBAND - ideogram track (DONE - 2012-10-19 - Hiram)
    ssh hgwdev
    mkdir -p /hive/data/outside/ncbi/ideogram/2012-10
    cd /hive/data/outside/ncbi/ideogram/2012-10
    # fetch all the ideogram files:
    rsync -a -P rsync://ftp.ncbi.nlm.nih.gov/pub/gdp/ ./

    mkdir /hive/data/genomes/mm10/bed/cytoband
    cd /hive/data/genomes/mm10/bed/cytoband

    # Create bed file
    $HOME/kent/src/utils/ncbi/createNcbiCytoBand.pl \
/hive/data/outside/ncbi/ideogram/2012-10/ideogram_10090_GCF_000000055.19_NA_V2

    ## can now verify before load:
    $HOME/kent/src/utils/ncbi/cytoBandVerify.pl
    #	everything checks out OK on 21 chroms
    # Load the bed file
    hgLoadBed -noBin -sqlTable=$HOME/kent/src/hg/lib/cytoBand.sql \
	mm10 cytoBand cytoBand.bed
    #   Read 403 elements of size 5 from cytoBand.bed
    # Make cytoBandIdeo track for ideogram gif on hgTracks page.
    # For mouse cytoBandIdeo is just a replicate of the cytoBand track.
    hgsql -e "drop table cytoBandIdeo;" mm10
    hgsql mm10 -e "create table cytoBandIdeo (index(chrom(10),chromStart)) as select * from cytoBand;"

##########################################################################
# CYTOBANDIDEO update -  (DONE - 2013-02-27 - kuhn)
# adding rows for chroms with no cytology
# this is just for navigation/orientation on those chroms

    set db=mm10
    set sql=~/kent/src/hg/lib/cytoBandIdeo.sql
    # make backup of existing table
    hgsql -e "CREATE TABLE cytoBandIdeoCopy SELECT * FROM cytoBandIdeo" $db
    # dump existing table
    hgsql -N -e "SELECT * FROM cytoBandIdeo" $db > $db.cytoBandIdeo

    # find chroms already covered
    hgsql -N -e 'SELECT chrom FROM cytoBandIdeo' $db \
       | sort -u > $db.coveredNames
    # make cytoBand records for chroms not already covered
    hgsql -N -e 'SELECT chrom, size FROM chromInfo' $db \
      | grep -wvf $db.coveredNames \
      | awk '{print $1"\t0\t"$2"\t\tgneg"}' > $db.cytoBandNew
    # check
    wc -l $db.*
    # combine and sort
    cat $db.cytoBandNew $db.cytoBandIdeo > $db.cytoBandIdeoFull
    bedSort $db.cytoBandIdeoFull $db.cytoBandIdeoFull
    # replace exsting table
    hgsql -e "DROP TABLE cytoBandIdeo" $db
    hgLoadSqlTab $db cytoBandIdeo $sql $db.cytoBandIdeoFull
    # check and then drop copy

##########################################################################
# lastz Lamprey petMar2 (DONE - 2012-10-17 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S petMar2
    mkdir /hive/data/genomes/mm10/bed/lastzPetMar2.2012-10-19
    cd /hive/data/genomes/mm10/bed/lastzPetMar2.2012-10-19

    cat << '_EOF_' > DEF
# Mouse vs. Lamprey
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz
BLASTZ_Y=3400
BLASTZ_L=6000
BLASTZ_K=2200
BLASTZ_M=50
BLASTZ_Q=/scratch/data/blastz/HoxD55.q

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000

# QUERY: Lamprey PetMar2
SEQ2_DIR=/hive/data/genomes/petMar2/petMar2.2bit
SEQ2_LEN=/hive/data/genomes/petMar2/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=60

BASE=/hive/data/genomes/mm10/bed/lastzPetMar2.2012-10-19
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
	-qRepeats=windowmaskerSdust \
        -chainMinScore=5000 -chainLinearGap=loose > do.log 2>&1 &
    #   real    218m29.078s

    cat fb.mm10.chainPetMar2Link.txt
    #   28262565 bases of 2652783500 (1.065%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzPetMar2.2012-10-19 lastz.petMar2

    #	and for the swap
    mkdir /hive/data/genomes/petMar2/bed/blastz.mm10.swap
    cd /hive/data/genomes/petMar2/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzPetMar2.2012-10-19/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
    #   real    7m2.754s
    cat  fb.petMar2.chainHg19Link.txt
    #	20923095 bases of 647368134 (3.232%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/petMar2/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# lastz White Rhino cerSim1 (DONE - 2012-10-23 - Hiram)
    # establish a screen to control this job with a name to indicate what it is
    screen -S mm10CerSim1
    mkdir /hive/data/genomes/mm10/bed/lastzCerSim1.2012-10-23
    cd /hive/data/genomes/mm10/bed/lastzCerSim1.2012-10-23

    cat << '_EOF_' > DEF
# Mouse vs. White Rhino
BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.02/bin/lastz

# TARGET: Mouse Mm10
SEQ1_DIR=/scratch/data/mm10/mm10.2bit
SEQ1_LEN=/scratch/data/mm10/chrom.sizes
SEQ1_CHUNK=20000000
SEQ1_LAP=10000
SEQ1_LIMIT=10

# QUERY: White Rhino CerSim1
SEQ2_DIR=/hive/data/genomes/cerSim1/cerSim1.2bit
SEQ2_LEN=/hive/data/genomes/cerSim1/chrom.sizes
SEQ2_CHUNK=10000000
SEQ2_LAP=0
SEQ2_LIMIT=20

BASE=/hive/data/genomes/mm10/bed/lastzCerSim1.2012-10-23
TMPDIR=/scratch/tmp
'_EOF_'
    # << happy emacs

    # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable
    #	number of jobs, 50,000 to something under 100,000
    # when not present, SEQ2_LIMIT is a default 100
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
        `pwd`/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
    #   real    992m45.890s

    cat fb.mm10.chainCerSim1Link.txt
    #   942281365 bases of 2652783500 (35.520%) in intersection
    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/mm10/bed
    ln -s lastzCerSim1.2012-10-23 lastz.cerSim1

    #	and for the swap
    mkdir /hive/data/genomes/cerSim1/bed/blastz.mm10.swap
    cd /hive/data/genomes/cerSim1/bed/blastz.mm10.swap
    time nice -n +19 doBlastzChainNet.pl -verbose=2 \
	/hive/data/genomes/mm10/bed/lastzCerSim1.2012-10-23/DEF \
        -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
        -swap -chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
    #   real    62m44s
    cat  fb.cerSim1.chainMm10Link.txt
    #	926131511 bases of 2366858012 (39.129%) in intersection

    # set sym link to indicate this is the lastz for this genome:
    cd /hive/data/genomes/cerSim1/bed
    ln -s blastz.mm10.swap lastz.mm10

#########################################################################
# QPCR PRIMERS (DONE - 2012-12-10 - Chin)
# The track name is changed to "qPCR Primers"
# Reload table with new track_mouse.BED (2013-01-28)
    # Download
    mkdir /hive/data/outside/Weizmann/qPcrPrimers
    cd /hive/data/outside/Weizmann/qPcrPrimers
    wget http://www.weizmann.ac.il/complex/compphys/software/Amit/primers/mouse/track_mouse.BED
    mkdir -p /hive/data/genomes/mm10/bed/qPcrPrimers
    cat track_mouse.BED | grep -v track \
     > /hive/data/genomes/mm10/bed/qPcrPrimers/qPcrPrimers_mm10.bed

    cd /hive/data/genomes/mm10/bed/qPcrPrimers
    hgLoadBed -bedDetail -tab -renameSqlTable \
      -sqlTable=$HOME/kent/src/hg/lib/bedDetail.sql \
      mm10 qPcrPrimers qPcrPrimers_mm10.bed
    # Reading qPcrPrimers_mm10.bed
    # Read 518230 elements of size 14 from qPcrPrimers_mm10.bed
    # Sorted
    # Creating table definition for qPcrPrimers
    # Saving bed.tab
    # Loading mm10

    # NULL descrition column
    hgsql mm10 -ne "UPDATE qPcrPrimers SET description = NULL;"

#########################################################################
# DBSNP B137 / SNP137 (DONE 12/20/12 angie)
# Redmine #7043

    mkdir -p /hive/data/outside/dbSNP/137/mouse
    cd /hive/data/outside/dbSNP/137/mouse
    # Look at the directory listing of ftp://ftp.ncbi.nih.gov/snp/database/organism_data/
    # to find the subdir name to use as orgDir below (mouse_10090 in this case).
    # Then click into that directory and look for file names like
    #    b(1[0-9][0-9])_*_([0-9]+_[0-9])
    # -- use the first num for build and the second num_num for buildAssembly.
    # jkStuff/liftContigs.lft maps NCBI contig names to chroms; use that for liftUp.
    #
    # Some trial and error was required to get the config.ra just right --
    # the b* filenames don't include buildAssembly!
    # patch contigs needed to be filtered out:
    cat > config.ra <<EOF
db mm10
orgDir mouse_10090
build 137
buildAssembly 
liftUp /hive/data/genomes/mm10/jkStuff/liftContigs.lft
EOF
    ~/kent/src/hg/utils/automation/doDbSnp.pl config.ra >& do.log & tail -f do.log
    # Script ended with feedback about needing refAssemblyLabel because dbSNP
    # mapped to more than one assembly; add the label that clearly corresponds to
    # mm10, GRCm38, to config.ra and try again:
    cat > config.ra <<EOF
db mm10
orgDir mouse_10090
build 137
buildAssembly 
liftUp /hive/data/outside/dbSNP/137/mouse/suggested.lft
refAssemblyLabel GRCm38
EOF
    ~/kent/src/hg/utils/automation/doDbSnp.pl -continue=loadDbSnp \
      config.ra >>& do.log & tail -f do.log
    # Script ended with feedback about unrecognized NT_* contigs from dbSNP.
    # Inspect the script-generated suggested.lft for liftUp; it's usually right.
    # For contigs that are labeled as part of GRCm38 but not liftable to mm10,
    # listed in script-generated cantLiftUpSeqNames.txt, do some entrez
    # nucleotide searches for contig IDs and convince yourself that they're all
    # for alt assembly sequences that we don't include in mm10 (e.g. patches,
    # other strains).  Then tell the script to filter out those contigs:
    cut -f 2 cantLiftUpSeqNames.txt > ignoreAltAssemblyContigs.txt
    cat > config.ra <<EOF
db mm10
orgDir mouse_10090
build 137
buildAssembly 
liftUp /hive/data/outside/dbSNP/137/mouse/suggested.lft
refAssemblyLabel GRCm38
ignoreDbSnpContigsFile /hive/data/outside/dbSNP/137/mouse/ignoreAltAssemblyContigs.txt
EOF
    ~/kent/src/hg/utils/automation/doDbSnp.pl -continue=loadDbSnp \
      config.ra >>& do.log & tail -f do.log
# ...
#MultipleAlignments      1667342 This variant aligns in more than one location.
#ObservedMismatch        4561144 UCSC reference allele does not match any observed allele from dbSNP.
#
# *** All done!
    # That is an unusually high count of ObservedMismatch... follow up with dbSNP.


#############################################################################
# FILTER SNP137 (DONE 12/21/12 angie)
    # Redmine #7043
    # Make several tracks that are filtered subsets of snp137:
    # First, filter out the multiply-aligned and/or weight >1 SNPs -> snp137Mult
    # Second, siphon off the common variants -> snp137Common
    # Third, take the (uniquely mapped, not known to be common) variants
    # w/dbSNP's "clinically-assoc" flag -> snp137Flagged
    cd /hive/data/outside/dbSNP/137/mouse
    zcat snp137.bed.gz \
    | perl -we \
      '$minTotal2N = 10; \
       ($multCount, $comCount, $flagCount, $miscCount) = (0,0,0,0); \
       open($mult, "| gzip -c > snp137Mult.bed.gz") || die; \
       open($common,    "| gzip -c > snp137Common.bed.gz") || die; \
       open($flagged,   "| gzip -c > snp137Flagged.bed.gz") || die; \
       open($misc,      "| gzip -c > snp137Misc.bed.gz") || die; \
       while (<>) { \
         @w = split("\t"); \
         if ($w[16] > 1 || $w[17] =~ /MultipleAlignments/) { \
           print $mult $_; \
           $multCount++; \
         } else { \
           my ($alleleFreqCount, $nStr, $freqStr) = ($w[20], $w[22], $w[23]); \
           my @alNs = split(",", $nStr);      die unless scalar(@alNs) == $alleleFreqCount; \
           my @freqs = split(",", $freqStr);  die unless scalar(@freqs) == $alleleFreqCount; \
           my ($total2N, $maxAlleleFreq) = (0, 0); \
           for (my $i = 0;  $i < $alleleFreqCount;  $i++) { \
             $total2N += $alNs[$i]; \
             $maxAlleleFreq = $freqs[$i] if ($freqs[$i] > $maxAlleleFreq); \
           } \
           if ($alleleFreqCount >= 2 && $total2N >= $minTotal2N && $maxAlleleFreq <= 0.99) { \
             print $common $_; \
             $comCount++; \
           } elsif($w[24] =~ /clinically-assoc/)  { \
             print $flagged $_; \
             $flagCount++; \
           } else { \
             print $misc $_; \
             $miscCount++; \
           } \
         } \
       } \
       close($mult);  close($common); close($flagged);  close($misc); \
       print "snp137Mult:    $multCount\nsnp137Common:  $comCount\nsnp137Flagged: $flagCount\n" . \
             "leftover:      $miscCount\n";'
#snp137Mult:    1671771
#snp137Common:  2709532
#snp137Flagged: 0
#leftover:      66537658
    # It's expected for snp137Flagged to be empty because that's for human SNPs.

    # Load tables
    foreach subset (Mult Common)
      hgLoadBed -tab -onServer -tmpDir=/data/tmp -allowStartEqualEnd -renameSqlTable \
        mm10 snp137$subset -sqlTable=snp137.sql snp137$subset.bed.gz
    end


############################################################################
# DBSNP CODING ANNOTATIONS (137) (DONE 12/21/12 angie)
    # Redmine #7043
    cd /hive/data/outside/dbSNP/137/mouse
    # ncbiFuncAnnotations.txt has NCBI coords: 0-based, fully closed.
    # For anything except an insertion (0 bases between flanks),
    # we need to add 1 to the end coord.  For an insertion, we need
    # to add 1 to the start coord.  Make a hash of the insertion IDs,
    # then look up each ID in ncbiFuncAnnotations.txt to tell which
    # transform to apply.
    # Note: sort -u with the keys below is too restrictive -- we need full line uniq.
    zcat ncbiFuncAnnotations.txt.gz \
    | perl -we 'open($IDS, "zcat ncbiFuncInsertions.ctg.bed.gz |") || die "ids: $!"; \
              while (<$IDS>) { chomp; $ids{$_} = 1; } \
              close($IDS); \
              %coding = (2=>1, 3=>1, 4=>1, 8=>1, 9=>1, 41=>1, 42=>1, 43=>1, 44=>1, 45=>1); \
              while (<>) { \
                chomp;  @w = split("\t"); # id, ctg, start, end, ... \
                next unless $coding{$w[5]}; \
                $bed4 = join("\t", $w[1], $w[2], $w[3], $w[0]); \
                if (exists $ids{$bed4} && $w[3] == $w[2]+1) { \
                  $w[2]++; # 2-base insertions: increment start coord \
                } else { \
                  $w[3]++; # increment end coord to get half-open \
                } \
                print join("\t", @w) . "\n"; \
              }' \
    | sort -k1n,1n -k2,2 -k3n,3n -k5,5 -k6n,6n \
    | uniq \
      > ncbiCodingAnnotations.txt
    wc -l ncbiCodingAnnotations.txt
#1884989 ncbiCodingAnnotations.txt

    # How many & what kinds of function types?
    cut -f 6 ncbiCodingAnnotations.txt \
    | sort -n | uniq -c
# 371388 3   (coding-synon)
#1301099 8   (cds-reference -- ignored)
#   3465 41  (nonsense)
# 199148 42  (missense)
#    319 43  (stop-loss)
#   7422 44  (frameshift)
#   2148 45  (cds-indel)
    # In b137, the functional annotations include non-coding (frame = NULL),
    # which we'll exclude here because this is supposed to be just coding stuff...
    # probably need to update how we show dbSNP's func annos anyway, e.g.
    # it is a shame that we toss out codon number and transcript offset.
    # Gather up multiple annotation lines into one line per {snp, gene, frame}:
    perl -e  'while (<>) { chomp; \
                my ($rsId, $ctg, $s, $e, $txId, $fxn, $frm, $nt, $aa, $codon) = split("\t"); \
                next if ($fxn == 8 && ($frm eq "NULL" && $aa eq "NULL" && $codon eq "NULL")); \
                if (defined $lastRs && \
                    ($lastRs != $rsId || $lastCtg ne $ctg || $lastS != $s || \
                     $lastTx ne $txId || $lastFrm ne $frm)) { \
                  if (defined $refRow) { \
                    $fxns = "$refRow->[0],$fxns";  $nts = "$refRow->[1],$nts"; \
                    $aas = "$refRow->[2],$aas";    $codons = "$refRow->[3],$codons"; \
                  } \
                  $lineOut = "$lastCtg\t$lastS\t$lastE\trs$lastRs\t$lastTx\t$lastFrm\t" . \
                        "$count\t$fxns\t$nts\t$codons\t$aas\n"; \
                  $lineOut =~ s@NULL@n/a@g; \
                  print $lineOut; \
                  $refRow = undef;  @rows = ();  ($count, $fxns, $nts, $codons, $aas) = (); \
                } \
                ($lastRs, $lastCtg, $lastS, $lastE, $lastTx, $lastFrm) = \
                    ($rsId, $ctg, $s, $e, $txId, $frm); \
                $count++; \
                if ($fxn == 8) { \
                  $refRow = [$fxn, $nt, $aa, $codon]; \
                } else { \
                 $fxns .= "$fxn,";  $nts .= "$nt,";  $aas .= "$aa,";  $codons .= "$codon,"; \
                } \
              } \
              if (defined $refRow) { \
                $fxns = "$refRow->[0],$fxns";  $nts = "$refRow->[1],$nts"; \
                $aas = "$refRow->[2],$aas";    $codons = "$refRow->[3],$codons"; \
              } \
              $lineOut = "$lastCtg\t$lastS\t$lastE\trs$lastRs\t$lastTx\t$lastFrm\t" . \
                    "$count\t$fxns\t$nts\t$codons\t$aas\n"; \
              $lineOut =~ s@NULL@n/a@g; \
              print $lineOut;' \
      ncbiCodingAnnotations.txt \
    | liftUp snp137CodingDbSnp.bed /hive/data/outside/dbSNP/137/mouse/suggested.lft warn stdin

    hgLoadBed mm10 snp137CodingDbSnp -sqlTable=$HOME/kent/src/hg/lib/snp125Coding.sql \
      -renameSqlTable -tab -notItemRgb -allowStartEqualEnd \
      snp137CodingDbSnp.bed
#Read 552120 elements of size 11 from snp137CodingDbSnp.bed


#########################################################################