Variance-quantitative trait loci enable systematic discovery of gene-environment interactions for cardiometabolic serum biomarkers

Abstract

Gene-environment interactions (GEIs) represent the modification of genetic effects by environmental exposures and are critical for understanding disease and informing personalized medicine. GEIs often induce differential phenotypic variance across genotypes; these variance-quantitative trait loci (vQTLs) can be prioritized in a two-stage GEI detection strategy to greatly reduce the computational and statistical burden and enable testing of a broader range of exposures. We performed genome-wide vQTL analysis for 20 serum cardiometabolic biomarkers by multi-ancestry meta-analysis of 350,016 unrelated participants in the UK Biobank, identifying 182 independent locus-biomarker pairs (p < 4.5×10−9). Most vQTLs were concentrated in a small subset (4%) of loci with genome-wide significant main effects, and 44% replicated (p < 0.05) in the Women’s Genome Health Study (N = 23,294). Next, we tested each vQTL for interaction across 2,380 exposures, identifying 846 significant GEIs (p < 2.4×10−7). Specific examples demonstrated interaction of triglyceride-associated variants with distinct body mass-versus body fat-related exposures as well as genotype-specific associations between alcohol consumption and liver stress at the ADH1B gene. Our catalog of vQTLs and GEIs is publicly available in an online portal.

Examining the evidence for Mendelian randomization homogeneity assumption violation using instrument association with exposure variance

“…Lack of an IV-exposure variance effect could suggest that the IV estimand estimates ACE subject to sufficient power to detect an effect. Identification of IV-exposure variance effects could enable follow up studies to identify the precise exposure-modifier interaction effect 13,14,16 . This could be useful to consider if this variable also modifies the exposure-outcome effect which would then imply NOSH assumption one is violated.…”

“…Testing for IV-exposure effect modification may be used as an empirical approach to detect violation of the IV homogeneity assumption 11 . Hypothesised testing of candidate IV-exposure interaction effects to evaluate homogeneity assumptions has been suggested 11 but this approach may miss unanticipated interaction effects, cannot be used if the modifier is unmeasured, and potentially incurs a large multiple testing burden 12,13 . Alternatively, the presence of effect modification can be identified by testing the association of the IV with exposure variance provided that the exposure is continuous [14][15][16] .…”