Biophysical Society Thematic Meeting| Santa Cruz 2018

Genome Biophysics: Integrating Genomics and Biophysics to Understand Structural and Functional Aspects of Genomes

Wednesday Speaker Abstracts

Complex Genetic Variants: Implications for Clinical Sequencing Methods and Validation Approaches Stephen Lincoln 1 ; Andrew Fellowes 2 ; Shazia Mahamdallie 3 ; Shimul Chowdhury 4 ; Eric Klee 5 ; Justin Zook 6 ; Rebecca Truty 1 ; Russell Garlick 7 ; Marc Salit 8 ; Nazneen Rahman 3 ; Stephen Kingsmore 4 ; Robert Nussbaum 1 ; Matthew Ferber 5 ; Brian Shirts 9 1 Invitae, San Francisco, California, United States; 2 Peter MacCallum Cancer Center, Melbourne, Victoria, Australia; 3 Institute of Cancer Research, London, United Kingdom; 4 Rady Children's Hospital, San Diego, California, United States; 5 Mayo Clinic, Rochester, Minnesota, United States; 6 NIST, Gaithersburg, Maryland, United States; 7 Seracare, Gaithersburg, Maryland, United States; 8 Stanford University, Palo Alto, California, United States; 9 University of Washington, Seattle, Washington, United States Objective: We evaluated the clinical prevalence of complex genetic variants and developed novel resources to help improve methods for detecting such variants. Background: DNA sequencing methods are often validated or benchmarked for their ability to detect simple single nucleotide variants (SNVs), small insertions and deletions (indels) and large copy number changes. Such variants are prevalent in every person, but most are not medically important. It is presently difficult to evaluate the performance of sequencing methodologies for more complex variant types. Such data are critical to determine which approaches may be most appropriate for specific medical tests. Methods: We analyzed over 80,000 patients undergoing clinical genetic testing using high-sensitivity methods. Guided by these data, a pilot specimen containing a diverse set of 22 challenging variants was engineered, validated, and provided to collaborating laboratories who sequenced it using 10 different NGS based workflows. Results: In our patient cohort, between 9 and 19% of the medically important (i.e. pathogenic) variants were of types that are technically challenging. These variants included small copy number changes, structural variants, large or complex indels, and repeat associated variants. Such variants were prevalent in genes critical to cancer risk assessment, cardiology, neurology, and pediatrics. In the interlaboratory study, most of the “easy” SNVs and indels in the pilot specimen were uniformly detected. However, only 10 of the 22 challenging variants were detected by all tests, and just 3 tests detected all 22. Many, but not all of these limitations were bioinformatic in nature and most were previously known but not well documented. Conclusions: The high prevalence of complex medically important variants is an under-recognized problem, incompletely addressed by current off-the-shelf DNA sequencing methods and control reagents. Approaches such as ours may help improve the standardization, quality, and transparency of clinical genetic tests.

23