Caspi et al. [2003] report that individuals who experience stressful life events (SLEs) have an increased risk of depression with each additional S allele, whereas for individuals who had never experienced SLEs, the S allele is not associated with depression. Despite support from animal studies [reviewed in Uher and McGuffin, 2008] attempts to replicate this G × E are also inconsistent. Large-scale studies that measure SLE and depression and collect DNA are costly in both time and money and their limited availability means most studies of this reported interaction are underpowered [Munafo et al., 2008]. Moreover, reviews of the same primary reports also reach different conclusions [Uher and McGuffin, 2008; Munafo et al., 2008]. Reported interactions comprise inconsistent modes of action (additive, S-dominant, and even L-dominant), crossover interactions and show evidence of publication bias [see also Willis-Owen et al., 2005; Uher and McGuffin, 2008; Munafo et al., 2008]. Munafo et al. [2008] conducted a meta-analysis of published results, but only five studies reported data in a format that allowed inclusion: they conclude “some claims of replication seem not to be
