Purchase this article with an account.
Terry Gaasterland, L. E. Edsall, A. D. Patel, S. A. Soares, R. C. Thompson, A. Y. Wu, D. E. Gaasterland, S. R. Head, P. L. Lee, R. Ayyagari; High-Fidelity Mapping Of Exome Sequences Improves Outcomes In Eye Disease Studies. Invest. Ophthalmol. Vis. Sci. 2011;52(14):3313.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Maximize discovery of real, eye-disease-related genome variations while minimizing false positive variant calls through DNA capture of exons genome-wide ("exome capture").
After consent, DNA was extracted from blood of individuals affected, or not affected, with one of several eye diseases. Commercial exome capture kits were applied to hybridize probes to target DNA in solution. For each person, captured DNA was sequenced (Illumina Genome Analyzer IIx or HiSeq) generating ~80 million 100-base reads. First, reads were mapped to a reference human genome (hg18; hg19) and SNPs (single nucleotide polymorphisms) identified using published parameters and methods. Second, to overcome deficiencies, we devised a new tiered strategy that maps perfectly matching reads and then iteratively relaxes match stringency. Exome SNPs were compared with known SNPs found independently with the more accurate, but low throughput, Sanger sequencing of the same samples.
Some of the exome SNPs were from pseudogenes. The tiered strategy mapped reads correctly to pseudogenes rather than genes, thus ensuring each exome SNP mapped to its true location. Further, this new tiered read mapping strategy partitioned reads into those that matched uniquely to exactly one position in the genome, or to two, to 3-5, or to 6-10. This allowed not only SNPs, but also insertions and deletions, to be predicted iteratively with strict subsets of mapped reads, and labeled according to mapping quality.
Tiered mapping and variant calling allows multi-mapped reads, which identify otherwise weakly supported genome variants. Such variations are identified as ambiguous and possibly resulting from gene conversion (a process by which pseudogene sequences are inserted into actual genes). With this method, in comparison with conventional mapping strategies, variant calls are ranked by the mapping quality of supporting reads. It also identifies possible gene conversion, increases sensitivity and specificity, and enhances potential to discover disease-related genes.
This PDF is available to Subscribers Only