The strengths of the current study include the large sample [only one other study, to date, is of similar size; Surtees et al., 2006] and accurate genotyping, including subdivision of the 5HTTLPR L allele according to the SNP rs25531. We use different outcome variables (ClinDep, suicidality and SelfRepDep) and, to interpret interactions in the context of scale of measurement, we use IRT analyses in addition to ordinal regressions. Finally, we addressed specific hypotheses that result from reviews of previous studies and limited our analyses to females, younger subjects, or limiting the time frame between stressful events and depression. Yet despite the strengths of the current study, and within the context of a strong association between PLE and depression/suicidality, we fail to replicate the hypothesized interaction [Caspi et al., 2003] between stress and the 5HTTLPR genotypes beyond that likely to occur by chance alone. Previous reports of replication should be considered within the context of potential problems including, poor quality genotyping, inconsistent types of interactions, inconsistent grouping of genotypes, selective presentation of results, interactions arising from the scale of measurement, and publication bias.
