Another interesting observation from GWAS has been the lack of consideration of the role of environment. As we describe below, we think GWAS may be limited by not examining how genetic influences on depression may vary among individuals with certain environmental exposures. One exception is a study by Powers and colleagues,57 who used propensity score matching to conduct a GWAS among case-control pairs matched on exposure to recent stressful life events. Use of propensity score matching enabled them to reduce sample heterogeneity and compare cases to controls with a similar level of exposure, though they did not formally test for GxE. In their analysis, no SNPs however, were genome-wide significant or even suggestive (p<5×10-6), though this was likely due to the very small sample size (n=805).
