GWAS of depressive symptoms have also been largely unrevealing. The first GWAS of depressive symptoms did not find any SNPs reaching genome-wide significance.55 One modestly associated (p=1.59×10-6) SNP (rs7582472) did show evidence of replication in two independent cohorts. However, this SNP was more than 300kb away from two genes and neither gene showed significant association to depression in a gene-based analysis. A second study of depressed mood also did not find any genome-wide significant SNP, but did find an intronic SNP (rs12912233) in RORA (retinoid related orphan receptor alpha gene) was modestly associated in the meta-analysis (p-6.3×10-7). While interesting, because another RORA SNP has been linked through GWAS to post-traumatic stress disorder,65 this result awaits replication. In the largest study, which was a meta-analysis comprising 17 population-based studies (n=34,549 individuals) as the discovery sample, no SNP reached genome-wide significance.56 The strongest association was for rs8020095 (p=1.05 × 10-7), located in the gene GPHN. When the discovery and replication samples were combined into one meta-analysis of 22 studies with 51,258 respondents, one region (indexed by the SNP rs40465) was associated with
