Identification of rare DNA variants in mitochondrial disorders with improved array-based sequencing

• Type of Material: Article
• Authors: Wenyi Wang, Stanford University; Peidong Shen, Stanford University; Sreedevi Thiyagarajan, Stanford University; Shengrong Lin, Stanford University; Curtis Palm, Stanford University; Rita Horvath, Newcastle University; Thomas Klopstock, University of Munich; David Cutler, Emory University; Lynn Pique, Stanford University; Iris Schrijver, Stanford University; Ronald W. Davis, Stanford University; Michael Mindrinos, Stanford University; Terence P. Speed, University of California Berkeley; Curt Scharfe, Stanford University
• Language: English
• Date: 2011-01-01
• Publisher: Oxford University Press (OUP): Policy C - Option B
• Publication Version: Final Published Version
• Copyright Statement: © 2010 The Author(s).
• License: Creative Commons Attribution-NonCommercial 2.5 Generic
• Final Published Version (URL): http://dx.doi.org/10.1093/nar/gkq750
• Title of Journal or Parent Work: Nucleic Acids Research
• ISSN: 0305-1048
• Volume: 39
• Issue: 1
• Start Page: 44
• End Page: 58
• Grant/Funding Information: National Eye Institute grant (R01EY016240 to R.W.D. and C.S.); T.K. is a member of MitoNET funded by the German Ministry of Education and Research (BMBF). Funding for open access charge: National Institutes of Health.
• Keywords: MUTATIONS; Science & Technology; MICROARRAYS; HIGH-THROUGHPUT; DELETIONS; GENOME; PROGRESSIVE EXTERNAL OPHTHALMOPLEGIA; ALGORITHM; Biochemistry & Molecular Biology; MODEL; POLYMORPHISMS; Life Sciences & Biomedicine; GENES
• Research Categories: Biology, Molecular; Chemistry, Biochemistry

Abstract

A common goal in the discovery of rare functional DNA variants via medical resequencing is to incur a relatively lower proportion of false positive base-calls. We developed a novel statistical method for resequencing arrays (SRMA, sequence robust multi-array analysis) to increase the accuracy of detecting rare variants and reduce the costs in subsequent sequence verifications required in medical applications. SRMA includes single and multi-array analysis and accounts for technical variables as well as the possibility of both low- and high-frequency genomic variation. The confidence of each base-call was ranked using two quality measures. In comparison to Sanger capillary sequencing, we achieved a false discovery rate of 2 (false positive rate 1.2×10-5, false negative rate 5), which is similar to automated second-generation sequencing technologies. Applied to the analysis of 39 nuclear candidate genes in disorders of mitochondrial DNA (mtDNA) maintenance, we confirmed mutations in the DNA polymerase gamma POLG in positive control cases, and identified novel rare variants in previously undiagnosed cases in the mitochondrial topoisomerase TOP1MT, the mismatch repair enzyme MUTYH, and the apurinic-apyrimidinic endonuclease APEX2. Some patients carried rare heterozygous variants in several functionally interacting genes, which could indicate synergistic genetic effects in these clinically similar disorders.