Large-scale multivariate multi-ancestry Interaction analyses point towards different genetic mechanisms by population and exposure

Abstract

The Gene-Lifestyle Interactions Working Group has recently conducted series of multi-ancestry genome-wide association screenings (GWAS) involving up to 610,475 individuals for three lipids (total triglyceride, high-density lipoprotein, and low-density lipoprotein), and four blood pressure traits (diastolic blood pressure, systolic blood pressure, mean arterial pressure, and pulse pressure) while accounting for potential interaction effect with drinking and smoking exposures. These GWAS reported both a 1 degree-of-freedom (df) test of gene-by-exposure (GxE) interaction, and a 2 df test of main genetic effect and interaction effect. Here we synthetized these results and generated a number of cross-studies statistics, providing a global perspective on this unique initiative, and suggesting guidelines for future GxE studies. We first summarized the signals identified across all phenotype-exposure pairs and ancestries considered, highlighting similarity and specificity of gene-environment interaction effects across phenotypes and ancestries. We then performed several follow-up analyses using genome-wide summary statistics, focusing in particular on multi-SNPs approaches and the relationship between interactions effect and marginal genetic effect. Those analyses identified new associations and demonstrated potential alternatives approaches for future GxE screenings. Finally, we explored differences in heritability conditional on the exposures considered. Our work provides new evidences for the potential implication of exposure-specific pathways, opening new paths for future gene-environment interaction studies.

Multi-omics insights into the biological mechanisms underlying statistical gene-by-lifestyle interactions with smoking and alcohol consumption

“…We obtained genetic summary statistics from genome-wide interaction analyses of cigarette smoking habits and alcohol consumption with lipid traits (HDL, LDL, and TG) and blood pressure traits (SBP, DBP, MAP, and PP) ( Table 1 ; Supplemental Table S1 ). The generation ( Feitosa et al, 2018 ; Sung et al, 2018 ; Bentley et al, 2019 ; de Vries et al, 2019 ; Sung et al, 2019 ) and harmonization ( Laville et al, 2020 ; Laville et al, 2022 ) of these summary statistics has been previously described, resulting in 140 sets of loci in four race/ancestry groups and one trans-ancestry meta-analysis ( Supplementary Figure S1 ).…”

“…A more complete discussion of these GLI association results are available in the primary publications for these projects. Briefly, some of the associations are for variants that are in higher frequency or present only among African ancestry populations, they are generally of low frequency (MAF 0.01–0.05) with high imputation scores, and with consistent associations across contributing African ancestry cohorts ( Sung et al, 2018 ; Bentley et al, 2019 ; Sung et al, 2019 ; Laville et al, 2020 ). Of the 30 loci prioritized in this study based on African ancestry meta-analyses, the lead associated variant was African ancestry-specific (only present in 1 KG AFR populations) for only 2, while for most the lead variants were available in all ancestries, but not associated in the meta-analyses of other ancestries.…”

Multi-omics insights into the biological mechanisms underlying gene-by-lifestyle interactions with smoking and alcohol consumption detected by genome-wide trans-ancestry meta-analysis

“…We used variant-level summary statistics from genome-wide gene-lifestyle interaction (GLI) analyses of cigarette smoking habits (current or ever smoking) and alcohol consumption (current or heavy vs. light drinking) with lipid traits (high-density lipoprotein cholesterol [HDL], low-density lipoprotein cholesterol [LDL], and triglycerides [TG]) and blood pressure traits (systolic blood pressure [SBP], diastolic BP [DBP], mean arterial pressure [MAP], and pulse pressure [PP]) ( Table 1; Supplemental Table 1 ). The generation(6-10) and harmonization(20) of these summary statistics has been previously described, resulting in 140 sets of results in four ancestry groups and one trans-ancestry meta-analysis ( Supplementary Figure 1 ). To restrict our efforts to regions in the genome most likely to have a GLI effect on the trait, we considered genetic variants with interaction p-value less than 5×10 −5 which resulted in 897 variants ( Figure 1 ).…”