GWAS signals revisited using human knockouts

Genome-wide association studies (GWAS) have been instrumental to our understanding of the genetic risk determinants of complex traits. A common challenge in GWAS is the interpretation of signals, which are usually attributed to the genes closest to the polymorphic markers that display the strongest statistical association. Naturally occurring complete loss of function (knockout) of these genes in humans can inform GWAS interpretation by unmasking their deficiency state in a clinical context. We exploited the unique population structure of Saudi Arabia to identify novel knockout events in genes previously highlighted in GWAS using combined autozygome/exome analysis. We report five families with homozygous truncating mutations in genes that had only been linked to human disease through GWAS. The phenotypes observed in the natural knockouts for these genes (TRAF3IP2, FRMD3, RSRC1, BTBD9, and PXDNL) range from consistent with, to unrelated to, the previously reported GWAS phenotype. We expand the role of human knockouts in the medical annotation of the human genome, and show their potential value in informing the interpretation of GWAS of complex traits.