The current findings are broadly consistent with earlier research published on a subset of the current data [Gillespie et al., 2005]. However, a criticism [Uher and McGuffin, 2008] of our earlier article [Gillespie et al., 2005] is that we failed to find an interaction, in part because there was only a weak association between SLE and depression (P = 0.08). The association in the current article is clearly stronger (P< 0.001), which in part is a result of the seven “problems getting along with” items being allocated as personal SLE, a decision based on a factor analysis of all SLE items (see Materials and Methods Section). In Gillespie et al. [2005] these items were included as network SLE. Nonetheless, our study does have limitations. Munafo et al. [2008] showed that the relationship between power and sample size is complex and depends on the size of the main effects and the proportion of individuals experiencing SLE. Based on our conservative power calculations, presented in Table II Supplementary, our largest sample (SelfRepDep) has substantial power to detect a G × E. However,
