GWAS OF PTSD IN >3M PARTICIPANTS IDENTIFIES HUNDREDS OF ASSOCIATIONS AND INSIGHTS INTO RARE VARIANT CONTRIBUTIONS

Background The Psychiatric Genomics Consortium for PTSD (PGC-PTSD) has assembled genetic data from over 3 million participants (> 500K cases), providing unprecedented power to identify common risk variation. In addition, to investigate how rare variants contribute to PTSD risk, we have assembled whole genome sequencing data of > 700K participants (100K cases). Here we present both the largest genome-wide association study (GWAS) and the first well-powered rare-variant association study (RVAS) of PTSD to identify both common and rare variants in PTSD. Methods Raw and summary data of 111 cohorts were contributed to PGC-PTSD. GWAS were performed using standardized PGC QC, imputation, and analysis protocols, including study-appropriate technical covariates. Sample size weighted fixed effects meta-analysis was performed using METAL, followed by post-GWAS analyses in FUMA. WGS data were quality controlled centrally by contributing biobanks. Variant annotations were performed using VEP, LOFTEE, and AlphaMissense. RVAS were performed using REGENIE with adjustment for 10 PCs, and meta-analyzed using REMETA. Results were stratified by functional category and multiple allele frequency thresholds were evaluated (from singletons to MAF < 1%) and combined using the GENE_P test. Results GWAS meta-analysis of PTSD identified > 300 genome-wide significant loci. The majority of associations are novel for PTSD, but previously identified loci, including genes such as FOXP2, MAD1L1, and ESR1 remain among the leading ones. Gene-based association analyses implicate > 500 genes, with an enrichment of genes in several pathways related to synapses (p < 1.25 × 10-6). RVAS identified multiple genes with function-altering variants modifying PTSD risk, including TET2 (p=7.4 × 10-21), ASXL1 (1.5 × 10-14) and DNMT3A (2.2 × 10-10) (gene-burden OR ranges 1.82-2.18). While gene-ranks differ for common and rare variant analyses, results are converging on mutual biological pathways. The largest GWAS and RVAS of PTSD have identified hundreds of risk loci and genes, highlighting synaptic processes among others in the etiology of PTSD. Additionally, given the role of gene x environment interactions via epigenetic mechanisms in PTSD, it is interesting that the top RVAS genes are epigenetic modifiers, with TET2 and DNMT3a as two of the best understood proteins that regulate DNA methylation. Together, findings provide the most complete picture of PTSD genetics to date.