Detecting Gene-Environment Interaction for Maternal Exposures Using Case-Parent Trios Ascertained Through a Case With Non-Syndromic Orofacial Cleft

Abstract

Two large studies of case–parent trios ascertained through a proband with a non-syndromic orofacial cleft (OFC, which includes cleft lip and palate, cleft lip alone, or cleft palate alone) were used to test for possible gene–environment (G × E) interaction between genome-wide markers (both observed and imputed) and self-reported maternal exposure to smoking, alcohol consumption, and multivitamin supplementation during pregnancy. The parent studies were as follows: GENEVA, which included 1,939 case–parent trios recruited largely through treatment centers in Europe, the United States, and Asia, and 1,443 case–parent trios from the Pittsburgh Orofacial Cleft Study (POFC) also ascertained through a proband with an OFC including three major racial/ethnic groups (European, Asian, and Latin American). Exposure rates to these environmental risk factors (maternal smoking, alcohol consumption, and multivitamin supplementation) varied across studies and among racial/ethnic groups, creating substantial differences in power to detect G × E interaction, but the trio design should minimize spurious results due to population stratification. The GENEVA and POFC studies were analyzed separately, and a meta-analysis was conducted across both studies to test for G × E interaction using the 2 df test of gene and G × E interaction and the 1 df test for G × E interaction alone. The 2 df test confirmed effects for several recognized risk genes, suggesting modest G × E effects. This analysis did reveal suggestive evidence for G × Vitamin interaction for CASP9 on 1p36 located about 3 Mb from PAX7, a recognized risk gene. Several regions gave suggestive evidence of G × E interaction in the 1 df test. For example, for G × Smoking interaction, the 1 df test suggested markers in MUSK on 9q31.3 from meta-analysis. Markers near SLCO3A1 also showed suggestive evidence in the 1 df test for G × Alcohol interaction, and rs41117 near RETREG1 (a.k.a. FAM134B) also gave suggestive significance in the meta-analysis of the 1 df test for G × Vitamin interaction. While it remains quite difficult to obtain definitive evidence for G × E interaction in genome-wide studies, perhaps due to small effect sizes of individual genes combined with low exposure rates, this analysis of two large case–parent trio studies argues for considering possible G × E interaction in any comprehensive study of complex and heterogeneous disorders such as OFC.

Benchmarking statistical methods for analyzing parent–child dyads in genetic association studies

“…Genotyping on the Illumina Human610 Quadv1_B array with 589,945 SNPs was performed at the Center for Inherited Disease Research (https://cidr.jhmi.edu/). As part of two recent publications (Ray et al, 2021, Zhang et al, 2021), trio‐aware phasing and reimputation using the 1000 Genomes Phase 3 release 5 reference panel were performed. Among GENEVA participants who were reimputed and used in these two articles, we restricted our analysis to the participants ascertained through nonsyndromic CL/P.…”