Purchase this article with an account.
Michael H. Farkas, Greg Grant, Maria Sousa, Eric A. Pierce; Transcriptome Analyses Identify Novel Exons In Known Inherited Retinal Disease Genes. Invest. Ophthalmol. Vis. Sci. 2012;53(14):1611.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Mutations in 186 genes have been identified to cause inherited retinal degeneration (IRD), accounting for 50-60% of cases. With the advent of next-generation genomic sequencing, many resources are focused on identifying novel mutations in the current annotation of coding sequence. However, RNA-seq transcriptome analyses have demonstrated that the current annotations are greatly under-representative and tissue-dependent. Here, we present the use of RNA-seq transcriptome analyses of the normal human neural retina to better characterize the coding sequence within known IRD genes.
Using RNA-seq, we generated approximately 100 million reads each from three normal human neural retina transcriptomes. We aligned the reads and analyzed the data using the RUM Unified Mapper (RUM) pipeline. The processed data was uploaded to the UCSC Genome Browser and the 186 currently known IRD genes were scanned to find novel exons. The frame of the novel exon was determined and validated using RT-PCR.
By analyzing 300 million reads, were able to obtain deep coverage of the normal human neural retina transcriptome, identifying nearly 80% of all annotated exons. This equates to full coverage of the expressed annotated human retinal transcriptome. However, we were able to identify nearly 25,000 additional novel exons. Within the set of known IRD genes, over 200 novel exons were identified in 99 genes. These novel exons represent alternate first and last exons, as well as internal exons. Using RT-PCR and Sanger sequencing, we validated novel exons in a subset of these transcripts including RP1 and MYO7A. All of the novel exons were confirmed to be in the spliced transcript.
The addition of the novel exons in the IRD genes account for a significant portion of all of the annotated exons in this set. Given that only 50-60% of IRD cases have been characterized, this finding has implications for finding new mutations that cause disease. On a larger scale, a better annotation of the current transcriptome will add nearly a megabase of additional coding sequence for screening of disease mutations in genes not previously characterized as disease-causing.
This PDF is available to Subscribers Only