Polygenic risk scores (PRS) were used to assess the predictive ability of the current genome-wide meta-analysis of depression within the clinically diagnosed MDD cohorts of Generation Scotland (GS; 975 cases and 5,971 controls), Münster (960 cases and 834 controls) and BiDirect (811 cases and 469 controls). PRS were also calculated using summary statistics from Wray, et al. 9 (Wray PRS) for comparison. Both the meta-analysis PRS and the Wray PRS were significantly associated with MDD in each of the three cohorts and the current meta-analysis PRS explained a greater proportion of the phenotypic variance in each target cohort compared to Wray PRS (Table 2 and Supplementary Figure 2).
