F20. INVESTIGATING THE SEX-SPECIFIC RELATIONSHIP BETWEEN POSTTRAUMATIC STRESS DISORDER AND MAJOR DEPRESSIVE DISORDER SYMPTOMS IN THE UK BIOBANK

• Published in: European neuropsychopharmacology 2024-10, Vol.87, p.215-216
• Main Authors: Moo-Choy, Ashley, Viola, Bence, Felsky, Daniel
• Format: Article
• Language: English
• ISSN: 0924-977X
• DOI: 10.1016/j.euroneuro.2024.08.431

Post-traumatic stress disorder (PTSD) and major depressive disorder (MDD) are known to have significant overlap in symptomatology. In accordance with their clinical similarities, genetic research into PTSD and MDD has identified overlap in heritable risk loci for both diagnoses. This overlap is imperfect, and we posit that PTSD and MDD require additional methods of genetic investigation to reveal the distinct etiologies that predispose individuals to each. Recent studies have worked to better disentangle their etiological distinctions but have largely relied on aggregate categorical diagnoses. We will instead use a symptom-specific approach, as genetic risk for specific PTSD and MDD symptoms and their interconnectedness have not yet been studied. Further, we know the experience of PTSD and MDD symptoms to be sex-specific and will therefore also perform sex-specific analyses to conclude whether differences exist between shared and distinct genetic etiology of PTSD and MDD symptoms by sex. The present study consists of two parts: (i) genetic correlation between symptoms of PTSD and MDD, and (ii) genomic structural equations modeling (gSEM) of PTSD and MDD symptoms, both using data from the UK Biobank (UKB; n = 157,366). The UKB sample includes the 9-item patient health questionnaire (PHQ-9) for measuring depression and the 6-item PTSD checklist (PCL-6). In Part I, correlation will be calculated, with results visualized using heat maps in R 4.4.0. In Part II, genome-wide association study (GWAS) results data for PTSD and MDD will be used to create structural equation models using PHQ-9 and PCL-6 symptom measures, which are able to visualize the joint genetic architecture between all 15 symptoms via their underlying factor structure in each sample. Results will include (i) multiple heat maps (total sample and sex-specific), with PHQ-9 measures on one axis and PCL-6 measures on the other, and (ii) corresponding structural equation models demonstrating the correlative relationship between symptom measures based on genetic information. While previous studies have demonstrated a high similarity between MDD and PTSD using genetic correlation, this has not yet been performed in the UKB using symptom-specific data. We may therefore compare the expected genetic correlation to those calculated through structural equation modelling using the same dataset. By comparing expected genetic correlation to those observed in our models, we can assess whether gSEM is capable of