QC statistics for lane 1

^⊥   Fraction of reference that is covered at least once   ^⊄   Estimated heterozygosity (average over multiplexes). Sample contamination can increase this estimate. Filters: base quality >= 39, mapping quality >= 30, pairs properly mapped, no indels, maximum 2 high-quality (Q30) variants in read, insert size > read length.   ^*   Mean coverage is computed over regions that are covered at least once   ^ℵ   Proportion of reads in regions with coverage in top 0.1 percentile   

Lane QC statistics and plots

G+C histogram∋		Insert size histogram∞		Mapped coverage by G+C℘
Coverage histogram		Exon/genome coverage distribution		Genomic coverage by G+C∅
(Predicted) variants by cycle∇ (read 1)		Fraction N/lowQ∋, read 1		G+C by cycle (PF)∋, read 1
Mean Q by cycle∇, read 1		Q score histogram, read 1		Variants by Q, read 1
(Predicted) variants by cycle∇ (read 2)		Fraction N/lowQ∋, read 2		G+C by cycle (PF)∋, read 2
Mean Q by cycle∇, read 2		Q score histogram, read 2		Variants by Q, read 2
Variants by GC		Indel rate by homopolymer content		Legend

^∋   The Fraction N/Low Q plots, and dotted lines on the GC histogram plots, refer to all reads that have passed chastity filters. If a reference genome was available, all others refer to mapped reads, otherwise they too refer to chastity-filtered reads. The dotted lines in the fraction N/lowQ plot correspond to the fraction of bases with quality score 4 or less.   ^∇   "Predicted variants" (dashed line) is the expected error frequency expressed as a Phred score, and may be compared with the "Variants by cycle" graph (solid line). "Mean Q" (solid) is the numerical mean Q score and is a measure of the average information content per read. These graphs use mapped reads only; the dashed line in the Mean Q plot uses all (PF) reads. All four graphs are calculated on called bases with Q Phred score above 4 only.   ^℘   Mapped coverage by G+C. The coverage was averaged over those genomic regions that were covered at least once. Regions with coverage in the top 0.1 percentile were excluded; the dotted line shows results for all reads. The G+C fraction was computed from read bases, excluding Ns and bases with quality below 4.   ^∅   Genomic coverage by G+C. The G+C fraction was computed from the reference genome, over the approximate fragment Regions with coverage in the top 0.1 percentile were excluded. The G+C histogram is shown as a dotted line (arbitrary Y scale).   ^∞   The insert size distribution is summarized by the median and median absolute deviation.   

Multiplex QC statistics

Yield (Mrd)		Median insert size		MAD insert size
Mean G+C		Std. dev. G+C by reads		Std. dev. G+C by position

Tile QC statistics and plots

Variant rate by tile∝ (read 1)		Raw/mapped yield by tile (read 1)		Fraction N/lowQ by tile (read 1)
Variant rate by tile∝ (read 2)		Raw/mapped yield by tile (read 2)		Fraction N/lowQ by tile (read 2)

^∝   HiSeq tiles are grouped in order: swathe 1 top; swathe 1 bottom; swathe 2 top; etc.