Only recently has substantial progress been made in understanding the underlying genetic architecture of depression, led by the Psychiatric Genomics Consortium (PGC) and a large meta-analysis combining results from PGC,7 UKB,8 FinnGen (http://r2.finngen.fi/pheno/F5_MOOD) and 23andMe.9,10 In this article, we describe genome-wide association analysis (GWAS) of ~310,000 participants from the U.S. Department of Veterans Affairs (VA) Million Veteran Program (MVP). MVP is one of the largest and most diverse biobanks in the world with genetic and electronic health record (EHR) data available. Several approaches have previously been taken regarding phenotypes selected for study for a depression GWAS. The PGC2 report7 used a variety of ascertainment methods within the cohorts used for meta-analysis, with a range of case definitions, including expert or clinician ascertainment of formal diagnostic major depressive disorder (MDD) criteria or treatment registers for approximately half of the cohorts, and combinations of self-report and clinical cutoffs on those self-report measures accounting for the other half.7 Other studies 8,10 investigated a broader trait definition of depression, which provided a larger sample size; a greater number of novel loci were discovered, with
