Chapter 11 - Copy Number Variant Detection Using Next-Generation Sequencing

Abstract

Copy number variation refers to imbalanced structural variation resulting in gain or loss of genomic DNA in a region ranging from tens to millions of bases long. Copy number variants (CNVs) are a critical class of genetic variation in clinical next-generation sequencing (NGS) due to their wide-ranging impacts in human disease. The variety and complexity of methods used to detect CNVs in NGS data can be daunting. This chapter will summarize general approaches to CNV detection from NGS data, application of such methods across diverse clinical NGS applications, and screening of CNVs to identify causal variants in human disease. We briefly describe the origins and the functional significance of CNVs and review non-NGS CNV detection technologies. Considerations of CNV screening across various clinical NGS contexts are discussed, including candidate gene, exome, whole genome, and cell-free DNA sequencing applications. The main conceptual approaches to CNV discovery in NGS data are reviewed in detail, including mate pair alignment, relative depth of coverage, and direct assembly or sequence-based evidence from split reads. CNV detection methods are highly design- and platform-dependent, requiring customization and trade-offs for specific approaches. We discuss the strengths and weaknesses of different CNV detection approaches in the context of both capture-based and whole genome sequencing platforms, highlighting considerations for germline DNA, tumor tissue, and cell-free tumor or fetal DNA. We discuss the availability of reference standards for benchmarking CNV detection by NGS methods. Finally, we cover orthogonal validation technologies toward the goal of incorporating clinically actionable NGS-based CNV detection for patient care.