Next-generation diagnostics: gene panel, exome, or whole genome?

Yu Sun,Claudia A L Ruivenkamp,Mariëtte J V Hoffer,Terry Vrijenhoek,Marjolein Kriek,Christi J van Asperen,Johan T den Dunnen,Gijs W E Santen
DOI: https://doi.org/10.1002/humu.22783
2015-01-01
Human Mutation
Abstract:Although the benefits of next-generation sequencing (NGS) for the diagnosis of heterogeneous diseases such as intellectual disability (ID) are undisputed, there is little consensus on the relative merits of targeted enrichment, whole-exome sequencing (WES) or whole-genome sequencing (WGS). To answer this question, WES and WGS data from the same nine samples were compared, and WES was shown not to miss any variants identified by WGS in a gene panel including approximate to 500 genes linked to ID (500GP). Additionally, deeply sequenced WES data were shown to adequately cover approximate to 99% of the 500GP; thus, little additional benefit was to be expected from a targeted enrichment approach. To reduce costs, minimal sequencing criteria were determined by investigating the relation between sequenced reads and outcome parameters such as coverage and variant yield. Our analysis indicated that 60 million reads yielded a mean coverage of approximate to 60x: approximate to 97% of the 500GP sequences were sufficiently covered to exclude variants, whereas variant yield was approximate to 99.5% and false-positive and false-negative rates were controlled. Our findings indicate that WES is currently the optimal approach to ID diagnostics. This result depends on the capture kit and sequencing strategy used. The developed framework however is amenable to other sequencing approaches.
What problem does this paper attempt to address?