Whole-exome sequencing identifies rare genetic variants associated with human plasma metabolites

Metabolite levels measured in the human population are endophenotypes for biological processes. We combined sequencing data for 3,924 (whole-exome sequencing, WES, discovery) and 2,805 (whole-genome sequencing, WGS, replication) donors from a prospective cohort of blood donors in England. We used multiple approaches to select and aggregate rare genetic variants (minor allele frequency [MAF] < 0.1%) in protein-coding regions and tested their associations with 995 metabolites measured in plasma by using ultra -high-performance liquid chromatography???tandem mass spectrometry. We identified 40 novel associations implicating rare coding variants (27 genes and 38 metabolites), of which 28 (15 genes and 28 metabolites) were replicated. We developed algorithms to prioritize putative driver variants at each locus and used mediation and Mendelian randomization analyses to test directionality at associations of metabolite and protein levels at the ACY1 locus. Overall, 66% of reported associations implicate gene targets of approved drugs or bioactive drug-like compounds, contributing to drug targets??? validating efforts.

Automated summary

Metabolite levels measured in the human population were analyzed to identify associations with rare genetic variants. The study identified 40 novel associations, of which 28 were replicated, between rare coding variants and metabolite levels. Algorithms were developed to prioritize driver variants and statistical analyses were used to test the directionality of associations between metabolite and protein levels. The study found that 66% of reported associations involved gene targets of approved drugs or bioactive drug-like compounds, validating efforts in drug target research.