View Program - 2022 Population, Evolutionary, and Quantitative Genetics Conference

Complex traits are influenced by many genetic variants that together determine the presentation and prevalence of phenotypes. How variants work together, additively or non-linear via epistasis, is unclear. If epistasis contributes to most complex diseases, an improved understanding of the complex architecture of complex traits will be critical to guide precision medicine-based decisions. The Hybrid Mouse Diversity Panel (HMDP) is a collection of over 100 inbred strains that can be used to detect and analyze genetic and environmental factors underlying complex traits. Previously reported HMDP phenotypic and genotypic data were analyzed in sex-stratified analyses for eight complex metabolic traits, to identify epistatic interactions. FaST-LMM, a linear mixed model method that accounts for population structure, measures single locus and two loci interaction associations, was used. In females, no marginal effects and 8 epistatic interactions in fat mass were detected using stringent Bonferroni multiple-testing correction. In males, significant marginal effects and interactions were detected for adiposity (58 marginal, 56,112 interactions), body weight (22 marginal, 66,296 interactions) and fat mass (60 marginal, 111,749 interactions). Interactions were also detected for cholesterol (57) and triglycerides (123). The relative number of detected main effects and interactions were comparable within each trait, as the number of pairwise interactions tested were more than 8,000 times greater than the number of main effects tested. Effect sizes and directions indicate that while the interactions were not significantly replicated between sexes at a Bonferroni adjusted level of significance, 80% or more of each trait’s interactions show the same direction of effect in the two sexes, significantly greater than the 50% expected by chance. Beyond the statistically significant interactions, there was also a significant enrichment in the number of marginal effects and interactions with a nominal p-value < 0.05 for all traits in both sexes, except for marginal effects on HDL in females. These results collectively demonstrate that the HMDP is a useful tool for detecting genetic architecture, especially interactions, even for traits that lack significant marginal effects. In conclusion, analysis of the HMDP enabled the discovery of marginal and interaction effects that may differ between sexes.

Sex-Dependency of Epistatic Interactions in the Hybrid Mouse Diversity Panel