This track shows RNA secondary structure predictions made with the EvoFold program, a comparative method that exploits the evolutionary signal of genomic multiple-sequence alignments for identifying conserved functional RNA structures.
Track elements are labeled using the convention ID_strand_score. At the zoomed-out level, secondary structure prediction regions are indicated by blocks, with the stem-pairing regions shown in a darker shade than unpaired regions. Arrows indicate the predicted strand. When zoomed in to the base level, the specific secondary structure predictions are shown in parenthesis format. The confidence score for each position is indicated in grayscale, with darker shades corresponding to higher scores.
The details page for each track element shows the predicted secondary structure (labeled SS anno), together with details of the multiple species alignments at that location. NOTE: This track is based on the ENCODE TBA 23-species alignment to the July 2003 human genome assembly, however the alignments displayed on the details page for this track are derived from the 8-way vertebrate alignment to the May 2004 human genome.
Substitutions relative to the human sequence are color-coded according to their compatibility with the predicted secondary structure (see the color legend on the details page). Each prediction is assigned an overall score and a sequence of position-specific scores. The overall score measures evidence for any functional RNA structures in the given region, while the position-specific scores (0 - 9) measure the confidence of the base-specific annotations. Base-pairing positions are annotated with the same pair symbol. The offsets are provided to ease visual navigation of the alignment in terms of the human sequence. The offset is calculated (in units of ten) from the start position of the element on the positive strand or from the end position when on the negative strand.
The graphical display may be filtered to show only those track elements with unnormalized scores that meet or exceed a certain threshhold. To set a threshhold, type the minimum score into the text box at the top of the description page.
Evofold makes use of phylogenetic stochastic context-free grammars (phylo-SCFGs), which are combined probabilistic models of RNA secondary structure and primary sequence evolution. The predictions consist of both a specific RNA secondary structure and an overall score. The overall score is essentially a log-odd score phylo-SCFG modeling the constrained evolution of stem-pairing regions and one which only models unpaired regions.
The predictions for this track were based on the conserved elements of the 23-way threaded blockset aligner (TBA) alignments present in the ENCODE regions (see the TBA Alignment track for more information).
The EvoFold program and browser track were developed by Jakob Skou Pedersen of the UCSC Genome Bioinformatics Group.
The 23-way TBA multiple alignments were created by Elliott Margulies of the Green Lab at NHGRI.
TBA was provided by Minmei Hou, Scott Schwartz and Webb Miller of the Penn State Bioinformatics Group.
Knudsen B, Hein J. RNA secondary structure prediction using stochastic context-free grammars and evolutionary history. Bioinformatics. 1999 Jun;15(6):446-54.
Pedersen JS, Bejerano G, Siepel A, Rosenbloom K, Lindblad-Toh K, Lander ES, Kent J, Miller W, Haussler D. Identification and classification of conserved RNA secondary structures in the human genome. PLoS Comput Biol. 2006 Apr;2(4):e33.
Pedersen JS, Meyer IM, Forsberg R, Simmonds P, Hein J. A comparative method for finding and folding RNA secondary structures within protein-coding regions. Nucl Acids Res. 2004 Sep 24;32(16):4925-36.
Siepel A, Bejerano G, Pedersen JS, Hinrichs AS, Hou M, Rosenbloom K, Clawson H, Spieth J, Hillier LW, Richards S, et al. Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res. 2005 Aug;15(8):1034-50.