Polygenic risk scores (PRS) were created using Plink v1.90b4 57 and evaluated in the Generation Scotland: the Scottish Family Health Study (GS) 60, Münster 9 and BiDirect cohorts 61 using the same method described in Wray, et al. 9. Two polygenic risk scores were created: one using weightings from the current meta-analysis and one using weightings from a previous association study of major depression conducted by Wray, et al. 9 (Wray PRS). To create independent SNP clumping was applied using a linkage disequilibrium r2 of < 0.1 and a 500kb sliding window. PRS were calculated using P-value thresholds of ≤ 5 × 10-8, ≤ 1 × 10-6, ≤ 1 × 10-4, ≤ 0.001, ≤0.01, ≤0.05, ≤0.1, ≤ 0.2, ≤ 0.5 and the full model including all SNPs (P ≤ 1). Nagelkerke’s R2 was used to calculate the proportion of phenotypic variance explained on the liability scale. PRS were split into deciles and the odds ratio for MDD in each decile calculated using the 1st decile as the reference group.
