Cross-ancestry genome-wide association study and systems-level integrative analyses implicate new risk genes and therapeutic targets for depression

Deciphering the genetic architecture of depression is pivotal for characterizing the associated pathophysiological processes and development of new therapeutics. Here we conducted a cross-ancestry genome-wide meta-analysis on depression (416,437 cases and 1,308,758 controls) and identified 287 risk loci, of which 140 are new. Variant-level fine-mapping prioritized potential causal variants and functional genomic analysis identified variants that regulate the binding of transcription factors. We validated that 80% of the identified functional variants are regulatory variants and expression quantitative trait loci (eQTL) analysis uncovered the potential target genes regulated by the prioritized risk variants. Gene-level analysis, including transcriptome-wide association study (TWAS), proteome-wide association study (PWAS), colocalization and Mendelian randomization-based analyses, prioritized potential causal genes and drug targets. Combining evidence from different analyses revealed likely causal genes, including TMEM106B, CTNND1, EPHB2, AREL1, CSE1L, RAB27B, SATU1, TMEM258, DCC, etc. Pathway analysis showed significant enrichment of depression risk genes in synapse-related pathways. Finally, we showed that Tmem106b knockdown resulted in depression-like behaviors in mice, supporting involvement of Tmem106b in depression. Our study identified new risk loci, likely causal variants and genes for depression, providing important insights into the genetic architecture of depression and potential therapeutic targets.