Data from the US Million Veteran Program (MVP) were used to replicate GWAS findings9. Participants reported here completed the PCL-C that asked respondents to report how much they have been bothered in the past 30 days by symptoms in response to stressful experiences (i.e. not just military experiences). The symptom cluster most distinctive for PTSD, re-experiencing symptoms (range 5–25), was analyzed. After accounting for missing phenotype data, the final sample for European Americans was 146,660, of whom 41.3% were combat-exposed. Genotyping was accomplished via a 723,305-SNP Affymetrix Axiom biobank array, customized for the MVP. Imputation was performed with Minimac 383 and the 1000 Genomes Phase 3 reference panel. GWAS analysis was conducted using RVTEST84 using linear regression with the first 10 principal components, age, and sex included as covariates. The results were filtered with imputation quality score R2 ≥ 0.9, MAF > 0.01 and HWE test P-value > 1 × 10-06. LDSC was used to estimate genetic correlation with the PGC2 EUA sample. The PGC2 EUA GWAS summary statistics were used to estimate PRS in MVP samples, where linear regression was then used to test for association between PRS and re-experiencing symptoms.
