PMC:2041973 / 36862-41860 JSONTXT

{"target":"http://pubannotation.org/docs/sourcedb/PMC/sourceid/2041973","sourcedb":"PMC","sourceid":"2041973","source_url":"http://www.ncbi.nlm.nih.gov/pmc/2041973","text":"Comparison of REAP to EST-Based Method and ACEScan\nTraditionally, AS exons were discovered by using EST alignments to genomic loci, and also more recently by computational algorithms that used sequence information extracted from multiple genomes. Here, we compared REAP predictions to both approaches. In the first comparison, publicly available ESTs and mRNA transcripts were aligned to the human genome sequence. 13,934 exons with evidence for exon-skipping and/or inclusion (EST-SE for EST-verified skipped exons) were generated, comprising ∼7% of all internal exons. First we analyzed Cyt-ES versus hCNS-SCns. If we required that none of the points per probeset (exon) was significant, 6% (4,402 of 71,731) of exons (after probeset mapping) had evidence for EST-SE (Figure 6A). Shuffling the mapping between these probesets and exons resulted in 8% (5,777 of 71,731) of exons with evidence for EST-SE (Figure 6A). These percentages were not significantly different from the 7% of exons with EST evidence for AS observed from using all exons. By raising the requirement that probesets had to contain at least one significant point to five significant points, the percentage of EST-SE increased dramatically from 11% (531 of 4,898 exons) to 26% (33 of 126). In comparison, the shuffled probesets at the same requirements remained at ∼8%, rising slightly to 11% at five points, due to small sample sizes. Similar trends were observed with hCNS-SCns versus HUES6-ES and the derived NPs versus hESCs (Figure 6A). Therefore, we concluded that REAP[+] exons were enriched for AS events independently identified by a transcript-based approach.\nFigure 6 Comparison of REAP Predictions for hCNS-SCns versus Cyt-hES, hCNS-SCns versus HUES6-ES, Cyt-NP versus Cyt-ES, and HUES6-NPs versus HUES6-ES with Alternative Exons Identified by an EST-Based Method and ACEScan\n(A) Black-filled squares represented the fraction of exons containing probesets with N significant points that had EST evidence for exon inclusion or exclusion (N = 0, 1, 2, 3, 4 and 5). White-filled triangles represented similarly computed fractions with permuted probeset to exon mappings.\n(B) Black-filled squares represented the fraction of exons containing probesets with N significant points that had ACEScan positive scores, indicative of evolutionarily conserved alternative exons. White-filled triangles represented similarly computed fractions with permuted probeset to exon mappings. Next, we compared REAP predictions to a computational approach of identifying exons with AS conserved in human and mouse, ACEScan [55]. ACEScan receives as input orthologous human–mouse exon pairs and flanking intronic regions and computes sequence features and integrates the features into a machine-learning algorithm to assign a real-valued score to the exon. A positive score indicated a higher likelihood of being AS in both human and mouse. ACEScan was updated in the following ways. Firstly, instead of relying on orthology information by Ensembl, and then aligning flanking introns in “orthologous” exons, conserved exonic and intronic regions in human and mouse from genome-wide multiple alignments were extracted. Secondly, whereas in our previous analysis exons from the longest transcript in Ensembl were utilized, now we collapsed all the transcripts available at the UCSC genome browser and analyzed all exons in the entire gene loci. ACEScan was utilized to assign ACEScan scores to all ∼162,000 internal exons in our genes. Exons annotated as first or last exons in Refseq mRNAs were excluded from our analysis, resulting in 4,487 positive-scoring exons, 2-fold more exons than originally published.\nHere we repeated our analysis with exons with positive ACEScan scores (ACE[+]) instead of EST-SEs. If we required that none of the points per probeset (exon) was significant, 2% (1,645 of 71,731) of exons (after probeset mapping) were ACE[+] (Figure 6B). Shuffling the mapping between these probesets and exons resulted in 3% (2,044 of 71,731) of exons being ACE[+] (Figure 6B). These percentages were not significantly different from the 2.7% observed from all exons (4,487 of the 162,000 exons that were scored by ACEScan). By raising the requirement that probesets had to contain five significant points, the percentage of ACE[+] exons increased from 4% to 11%. However, the sample sizes were small. In comparison, the shuffled probesets at the same requirements remained at ∼4%. Similar overall trends were observed with hCNS-SCns versus HUES6-ES and the derived NPs versus hESCs (Figure 6B). In total, 7.5% (131 of 1,737) of REAP [+] exons were designated as ACEScan[+] compared to 2.4% (2,328 of 97,437) of REAP[−] exons. This result suggested that a small but significantly enriched fraction of AS events in hESCs versus NPs was likely to be evolutionarily conserved in human and mouse. In conclusion, our results suggested that REAP predictions were congruent with predictions from two independent, orthogonal methods.","divisions":[{"label":"Title","span":{"begin":0,"end":50}},{"label":"Figure caption","span":{"begin":1640,"end":2455}},{"label":"Title","span":{"begin":1650,"end":1858}}],"tracks":[{"project":"2_test","denotations":[{"id":"17967047-15708978-85283466","span":{"begin":2587,"end":2589},"obj":"15708978"},{"id":"T61565","span":{"begin":2587,"end":2589},"obj":"15708978"}],"attributes":[{"subj":"17967047-15708978-85283466","pred":"source","obj":"2_test"},{"subj":"T61565","pred":"source","obj":"2_test"}]},{"project":"CellFinder","denotations":[{"id":"T293","span":{"begin":1764,"end":1773},"obj":"CellType"},{"id":"T294","span":{"begin":1752,"end":1758},"obj":"CellLine"},{"id":"T295","span":{"begin":1711,"end":1720},"obj":"CellType"},{"id":"T296","span":{"begin":1702,"end":1709},"obj":"CellLine"},{"id":"T297","span":{"begin":1781,"end":1789},"obj":"CellLine"},{"id":"T298","span":{"begin":1728,"end":1736},"obj":"CellLine"},{"id":"T299","span":{"begin":1711,"end":1718},"obj":"CellType"},{"id":"T300","span":{"begin":1685,"end":1692},"obj":"CellType"},{"id":"T301","span":{"begin":1728,"end":1733},"obj":"CellLine"},{"id":"T302","span":{"begin":1685,"end":1694},"obj":"CellType"},{"id":"T303","span":{"begin":1764,"end":1769},"obj":"CellLine"},{"id":"T304","span":{"begin":1781,"end":1786},"obj":"CellLine"}],"attributes":[{"subj":"T293","pred":"source","obj":"CellFinder"},{"subj":"T294","pred":"source","obj":"CellFinder"},{"subj":"T295","pred":"source","obj":"CellFinder"},{"subj":"T296","pred":"source","obj":"CellFinder"},{"subj":"T297","pred":"source","obj":"CellFinder"},{"subj":"T298","pred":"source","obj":"CellFinder"},{"subj":"T299","pred":"source","obj":"CellFinder"},{"subj":"T300","pred":"source","obj":"CellFinder"},{"subj":"T301","pred":"source","obj":"CellFinder"},{"subj":"T302","pred":"source","obj":"CellFinder"},{"subj":"T303","pred":"source","obj":"CellFinder"},{"subj":"T304","pred":"source","obj":"CellFinder"}]}],"config":{"attribute types":[{"pred":"source","value type":"selection","values":[{"id":"2_test","color":"#93c2ec","default":true},{"id":"CellFinder","color":"#dcec93"}]}]}}