GEM: scalable and flexible gene–environment interaction analysis in millions of samples

• DOI: 10.1093/bioinformatics/btab223
• Publication Date: 2021-04-07T19:56:19Z
• Authors: Kenneth E Westerman; Duy T Pham; Liang Hong; Ye Chen; Magdalena Sevilla...; Yun Ju Sung; Yan V Sun; Alanna C Morrison; Han Chen; Alisa K Manning
• Keywords: Male; 0301 basic medicine; Data Interpretation; Statistical; Original Papers; 004; 3. Good health; 03 medical and health sciences; Phenotype; Data Interpretation, Statistical; Medicine and Health Sciences; Humans; Female; Gene-Environment Interaction; Software; Genome-Wide Association Study

Abstract Motivation Gene–environment interaction (GEI) studies are a general framework that can be used to identify genetic variants modify the effects of environmental, physiological, lifestyle or treatment on complex traits. Moreover, accounting for GEIs enhance our understanding architecture diseases and However, commonly statistical software programs GEI either not applicable testing certain types hypotheses have been optimized use in large samples. Results Here, we develop new program, GEM (Gene–Environment analysis Millions samples), which supports inclusion multiple terms, adjustment covariates robust inference, while allowing multi-threading reduce computation time. conduct tests as well joint main both continuous binary phenotypes. Through simulations, demonstrate scales millions samples addressing limitations existing programs. We additionally gene-sex waist-hip ratio 352 768 unrelated individuals from UK Biobank, identifying 24 novel loci test previously reported combined sex-specific analyses. Our results facilitate next generation large-scale help advance Availability implementation is freely available an open source project at https://github.com/large-scale-gxe-methods/GEM. Supplementary information data Bioinformatics online.