Protein coding genes were mapped if they were located within 10Kb up- or downstream from index variants or if a credible variant was annotated to the gene based on eQTL data or chromatin interaction data from human brain (data sets used in the mapping can be found in Supplementary Table S7A). No additional variant filtering by functional annotation was applied in the eQTL and chromatin interaction mapping. This analysis identified 411 depression risk genes, which were used in a gene-set enrichment analysis within the GENE2FUNC module of FUMA, where we analyzed 8,664 gene-sets derived from GO Biological Process (N=7,658; 393 gene-sets) and GO Cellular Components (N=1,006; 86 gene-sets) ontology in the MSigDB database.
