The three studies5, 8, 9 contributing to the meta-analysis were based on different depression phenotypes; nevertheless, there were strong genetic correlations (> 0.85) between them. PGC_139k used a predominately clinically ascertained diagnosis of major depression and the observed genetic correlations indicate that each study was likely to be capturing a similar underlying genetic architecture, despite the use of different diagnostic instruments. Therefore, larger population-based samples, where the timescales and costs of obtaining a clinically diagnosis would be prohibitive, can contribute to our understanding of the genetic architecture of the disease. The meta-analysis also provides the opportunity to assess the general directionality agreement of variants that have previously been associated with depression (Supplementary Note and Supplementary Table 2). The examination of previous findings between studies and the current meta-analysis suggests that there is good degree of directionality agreement of causal effects of depression when using studies with over 100,000 participants. However, there is likely to be a need to continue to ascertain clinically ascertained MDD cohorts to ensure validity of the larger cohorts with broader phenotyping.
