Assessing the contribution of rare-to-common protein-coding variants to circulating metabolic biomarker levels via 412,394 UK Biobank exome sequences

Abstract

Genome-wide association studies have established the contribution of common and low frequency variants to metabolic biomarkers in the UK Biobank (UKB); however, the role of rare variants remains to be assessed systematically. We evaluated rare coding variants for 198 metabolic biomarkers, including metabolites assayed by Nightingale Health, using exome sequencing in participants from four genetically diverse ancestries in the UKB (N=412,394). Gene-level collapsing analysis, that evaluated a range of genetic architectures, identified a total of 1,303 significant relationships between genes and metabolic biomarkers (p<1x10-8), encompassing 207 distinct genes. These include associations between rare non-synonymous variants in GIGYF1 and glucose and lipid biomarkers, SYT7 and creatinine, and others, which may provide insights into novel disease biology. Comparing to a previous microarray-based genotyping study in the same cohort, we observed that 40% of gene-biomarker relationships identified in the collapsing analysis were novel. Finally, we applied Gene-SCOUT, a novel tool that utilises the gene-biomarker association statistics from the collapsing analysis to identify genes having similar biomarker fingerprints and thus expand our understanding of gene networks.

Rare and common genetic determinants of metabolic individuality and their effects on human health

“…Both networks (along with details for all association results) can be explored on our webserver at https://omicscience.org/apps/mgwas. metabolites) using nuclear magnetic resonance (NMR) 2,3 , but only a few, smaller-scale studies (N max ≈ 8,000) have been conducted using the much broader metabolite coverage of untargeted methods (up to 644 metabolites), which have each reported fewer than 150 loci 4,5 . In this Article we present a systematic investigation of the genetic architecture of over 900 metabolites in almost 20,000 men and women.…”

Whole Exome Sequencing Enhanced Imputation Identifies 85 Metabolite Associations in the Alpine CHRIS Cohort

“…Even though previous metabolite GWAS have benefited from a larger sample size or more metabolites tested, our study queries the entire coding space, which, due to the high cost of sequencing is infeasible for many mGWAS studies, who often use imputed genotype data [ 5 , 6 , 7 , 8 ]. Despite the smaller size of our panel compared to others, the Biocrates kit used here measures some metabolites missed by other common platforms, such as Metabolon [ 5 , 6 , 8 , 29 , 30 ] or Nightingale [ 6 , 11 ]. For example, Nightingale does not measure the amino acids tryptophan, arginine, or aspartate, and neither Metabolon nor Nightingale measures the biogenic amines histamine, spermine, or putrescine, or the acylcarnitines decadienylcarnitine or dodecanedioylcarnitine.…”

“…Among our significant findings, 15 rare associations are more than five times more common in CHRIS than in non-Finnish Europeans in gnomAD. This demonstrates the value of both exome sequencing followed by imputation for ExWAS and the value of exploring diverse population cohorts, even if they might be smaller compared to others, for the identification of novel variant-trait associations in the age of biobank-level studies [ 11 ].…”

Whole-exome sequencing study identifies novel rare variants and genes associated with intraocular pressure and glaucoma

“…Genetic variants with P < 1 × 10 −8 were declared ExWAS significant 36 . In addition to using European subjects, recent ExWAS studies advocate to include subjects of all ancestries 37,38 . Hence, we performed both white only and pan-ancestry analyses (added an additional covariate for four major ancestral groups, i.e., European, South Asian, East Asian, and African, identified by the K-Means clustering algorithm).…”