Explorations to improve the completeness of exome sequencing. Issue 1 (December 2016)

• Record Type: Journal Article
• Authors: Du, Chen; Pusey, Barbara; Adams, Christopher; Lau, C.; Bone, William; Gahl, William; Markello, Thomas; Adams, David
• Is Part Of: BMC medical genomics. Volume 9:Issue 1(2016)
• Page Start: 1
• Page End: 11
• Publication Date: 2016-12
• Subjects: Clinical exome sequencing -- Analytical quality -- Performance enhancement -- Clinical genomics -- Rare diseases -- Completeness problem -- False negative results; Medical genetics -- Periodicals; Genomics -- Periodicals; 616.042
• DOI: 10.1186/s12920-016-0216-3
• Languages: English
• ISSNs: 1755-8794

Abstract

Abstract Background Exome sequencing has advanced to clinical practice and proven useful for obtaining molecular diagnoses in rare diseases. In approximately 75 % of cases, however, a clinical exome study does not produce a definitive molecular diagnosis. These residual cases comprise a new diagnostic challenge for the genetics community. The Undiagnosed Diseases Program of the National Institutes of Health routinely utilizes exome sequencing for refractory clinical cases. Our preliminary data suggest that disease-causing variants may be missed by current standard-of-care clinical exome analysis. Such false negatives reflect limitations in experimental design, technical performance, and data analysis. Results We present examples from our datasets to quantify the analytical performance associated with current practices, and explore strategies to improve the completeness of data analysis. In particular, we focus on patient ascertainment, exome capture, inclusion of intronic variants, and evaluation of medium-sized structural variants. Conclusions The strategies we present may recover previously-missed, disease causing variants in second-pass exome analysis. Understanding the limitations of the current clinical exome search space provides a rational basis to improve methods for disease variant detection using genome-scale sequencing techniques.