Another possibility for these divergent results is that utilizing a GWAS approach to characterize genetic risk for alcohol problems may preclude identification of genetic effects that may be contingent upon exposure to environmental conditions (Caspi et al., 2010). For example, rs1229984 (Arg48His) on ADH1B has been identified and subsequently replicated in GWAS for its association with alcohol dependence across different populations (Bierut et al., 2012; Gelernter et al., 2014). Despite a robust main effect of rs1229984 on alcohol-related outcomes across samples, GxE studies involving this variant have so far produced inconsistent results (e.g., Olfson et al., 2014; Meyers et al., 2013; Sartor et al., 2014). Although this is partly attributable to differences in how studies measure the environment, robust genotype-to-phenotype associations detected in GWAS may not guarantee robust GxE findings. One possible direction for future studies is to characterize genetic variation in phenotypes that are plausibly related to how an individual reacts to environmental factors (e.g., stress sensitivity, social vulnerability) (Caspi et al., 2010).
