F2. genetic predictors of post-traumatic stress symptoms in the COVID-19 era

Background Post-traumatic stress disorder (PTSD) is a common and debilitating mental health condition. Although trauma exposure is widespread, only an estimated 5.6% of trauma-exposed individuals will develop symptoms, suggesting that individual differences in vulnerability — including genetic factors — play an important role. PTSD is known to be heritable and polygenic, and polygenic risk scores (PRS) offer a means of aggregating common genetic variants to quantify an individual's genetic liability. Most research examining the predictive utility of PTSD-PRS has focused on specialised, high-trauma populations, with limited investigation in the general population or in relation to post-traumatic stress symptoms (PTSS) more broadly. The COVID-19 pandemic represented a significant population-level stressor associated with elevated rates of PTSD and PTSS, yet the extent to which genetic risk interacted with pandemic-specific factors to influence symptom severity remains unclear. The South Australian experience was unique with the state effectively free of community COVID-19 transmission until November 2021 when borders were opened, presenting the opportunity to evaluate impact of pandemic timepoint as well as COVID-19 infection. Methods The Neurocognition and Genomics study collected participants from 2020 to 2024. Analyses were conducted in participants who provided a saliva sample and had complete survey data (n = 1,318; mean age = 48.9 years; 68% female). A PTSD-PRS was calculated using SBayesR, a Bayesian regression approach that estimates posterior SNP effect sizes while accounting for linkage disequilibrium, applied to summary statistics from the Psychiatric Genomics Consortium (Nievergelt et al., 2024). PTSS were assessed using the PTSD Checklist for DSM-5 (PCL-5), with probable PTSD defined as a score ≥31. Two COVID-19-related moderators were examined: SARS-CoV-2 infection status (determined via self-report) and pandemic timepoint, operationalised relative to the reopening of South Australian borders. Covariates included age, gender, and ancestry principal components. Results PTSD-PRS was a significant predictor of PTSS after covariate adjustment, explaining approximately 1% of the variance (F(1, 1309) = 15.96, FDRp = .004, η² = .01), with the total model accounting for 8% of variance (R² = .08, F(8, 1309) = 15.09, p < .001). Higher PTSD-PRS was associated with increased likelihood of meeting threshold for probable PTSD in both continuous PRS (χ²(11) = 59.25, p < .001, Nagelkerke R² =.08) and categorical quintile PRS (χ²(11) = 61.85, p < .001, Nagelkerke R² =.08) models. No significant moderating effects were observed for either SARS-CoV-2 infection status (F(2, 1192) = 0.40, p = .673) or pandemic timepoint (F(1, 1300) = 0.05, p = .820). Discussion These findings demonstrate that PTSD-PRS significantly predicts PTSS in a general population sample, supporting the utility of polygenic approaches outside of specialised trauma cohorts. Although no evidence was found for moderation by COVID-19-related factors, this work represents a novel integration of genomic and pandemic data. It should be noted that this sample may not be fully representative of the general population during the COVID-19 period, as those who attended in-person testing to provide DNA showed lower DASS and PTSD scores relative to those who did not. Strengthening polygenic understanding of PTSS and PTSD may help support better public health outcomes in the context of future pandemics.