Large-scale RACE approach for proactive experimental definition of C. elegans ORFeome
- Kourosh Salehi-Ashtiani1,
- Chenwei Lin1,
- Tong Hao1,
- Yun Shen1,
- David Szeto1,
- Xinping Yang1,
- Lila Ghamsari1,
- Hanjoo Lee1,
- Changyu Fan1,
- Ryan Murray1,
- Stuart Milstein1,
- Nenad Svrzikapa1,
- Michael Cusick1,
- Frederick Roth2,
- David Hill1 and
- Marc Vidal1,3
- * Corresponding author; email: marc_vidal{at}dfci.harvard.edu
Abstract
Although a highly accurate sequence of the C. elegans genome has been available for ten years, the exact transcript structures of many of its protein coding genes remain unsettled. Approximately two-thirds of the ORFeome has been verified reactively by amplifying and cloning computationally predicted transcript models; still a full third of the ORFeome remains experimentally unverified. To fully identify the protein coding potential of the worm genome including transcripts that may not satisfy existing heuristics for gene prediction, we developed a computational and experimental platform adapting Rapid Amplification of cDNA Ends (RACE) for large-scale structural transcript annotation. We interrogated two thousand unverified protein coding genes using this platform. We obtained RACE data for approximately two-thirds of the examined transcripts and reconstructed ORF and transcript models for close to one thousand of these. We defined untranslated regions, identified new exons, and redefined previously annotated exons. Our results show that as much as 20% of the C. elegans genome may be incorrectly annotated. Many annotation errors could be corrected proactively with our large-scale RACE platform.
Footnotes
-
- Received July 19, 2009.
- Accepted September 22, 2009.
- Copyright © 2009, Cold Spring Harbor Laboratory Press
