The PLINK programme (http://pngu.mgh.harvard.edu/purcell/plink) was used to determine Hardy–Weinberg equilibrium, and for genetic association analyses. Additive, dominant and recessive models were all tested with age and gender as covariates in all models. In case of interactional models, the main effect of the genotype and that of the life event variable were also included as covariates. Post hoc tests of all models were carried out in the two genders to establish gender differences. Nominal significance threshold was P<0.05, and we calculated false discovery rates (FDR) for the main tests by using the QVALUE programme http://github.com/jdstorey/qvalue. As we have large samples (n>200), parametric statistical methods were applied based on the central limit theorem.43 SPSS 20.0 (IBM, Armonk, NY, USA) for Windows was used for other statistical analyses, and to generate the figures.
