Re-analysis and meta-analysis of summary statistics from gene-environment interaction studies

Authors: PHAM, DUY - University Of Texas Health Science Center; WESTERMAN, KENNETH - Massachusetts General Hospital; PAN, CONG - University Of Texas Health Science Center; CHEN, LING - Massachusetts General Hospital; SRINIVASAN, SHYLAJA - University Of California San Francisco (UCSF); ISGANAITIS, ELVIRA - Joslin Diabetes Center; VAJRAVELU, MARY ELLEN - University Of Pittsburgh School Of Medicine; BACHA, FIDA - Children'S Nutrition Research Center (CNRC); CHERNAUSEK, STEVE - University Of Oklahoma; GUBITOSI-KLUG, ROSE - Case Western Reserve University (CWRU); DIVERS, JASMIN - New York University; PIHOKER, CATHERINE - University Of Washington Medical School; MARCOVINA, SANTICA - University Of Washington; MANNING, ALISA - Massachusetts General Hospital; CHEN, HAN - University Of Texas Health Science Center

Submitted to: Oxford Bioinformatics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/30/2023
Publication Date: 12/1/2023
Citation: Pham, D.T., Westerman, K.E., Pan, C., Chen, L., Srinivasan, S., Isganaitis, E., Vajravelu, M., Bacha, F., Chernausek, S., Gubitosi-Klug, R., Divers, J., Pihoker, C., Marcovina, S.M., Manning, A.K., Chen, H. 2023. Re-analysis and meta-analysis of summary statistics from gene-environment interaction studies. Bioinformatics. 39(12):Article btad730. https://doi.org/10.1093/bioinformatics/btad730.
DOI: https://doi.org/10.1093/bioinformatics/btad730

Interpretive Summary: Gene–environment interaction (GEI) analysis is a statistical analysis method to understand genetic impacts on human disease, while also accounting for additional the exposures in the environment. This can result in better understand the differences in genetic effects across populations, and support personalized lifestyle and therapeutic decisions. Researchers in Houston collaborated with statisticians to analyze data from a national multicenter cohort from the diabetes-focused ProDiGY consortium as well as from the United Kingdom biobank. These analyses provide new tools to investigate how the environment may influence the risk for disease while also taking into account genetic risk factors.

Technical Abstract: Summary statistics from genome-wide association studies enable many valuable downstream analyses that are more efficient than individual-level data analysis while also reducing privacy concerns. As growing sample sizes enable better-powered analysis of gene–environment interactions, there is a need for gene–environment interaction-specific methods that manipulate and use summary statistics. We introduce two tools to facilitate such analysis, with a focus on statistical models containing multiple gene–exposure and/or gene–covari-ate interaction terms. REGEM (RE-analysis of GEM summary statistics) uses summary statistics from a single, multi-exposure genome-wide interaction study to derive analogous sets of summary statistics with arbitrary sets of exposures and interaction covariate adjustments. METAGEM META-analysis of GEM summary statistics) extends current fixed-effects meta-analysis models to incorporate multiple exposures from multiple studies. We demonstrate the value and efficiency of these tools by exploring alternative methods of accounting for ancestry-related population stratification in genome-wide interaction study in the UK Biobank as well as by conducting a multi-exposure genome-wide interaction study meta-analysis in cohorts from the diabetes-focused ProDiGY consortium. These programs help to maximize the value of summary statistics from diverse and complex gene–environment interaction studies.