It has long been appreciated that many common diseases are influenced by genetic and environmental factors. Early adversity and stress later in life are potent predictors of a broad range of health risk behaviors and multiple medical problems. This proposal is being submitted in response to PAR-13-382, Analysis of Genome-Wide Gene-Environment (GxE) Interactions, which supports secondary analyses of Genome Wide Association Study (GWAS) datasets to examine GxE interactions. This application proposes to conduct secondary analyses using data from the Army Study to Assess Risk and Resilience in Service members (Army STARRS). This study (N=16,000) was designed to assess genomic predictors of Posttraumatic Stress Disorder (PTSD) and suicidality in a representative cohort of soldiers, and includes measures of early childhood adversity and adult traumatic events; GWAS, exome, and custom array genotyping data; smoking and alcohol use information, and a wealth of other health outcomes available via links with the Army and Department of Defense Medical Data Repository. The goal of this proposal is to identify genetic and GxE predictors of health risk behaviors that are associated with significant morbidity and mortality. Predictors of nicotine and alcohol dependence will be examined as primary aims, and indices of obesity (e.g., BMI, cholesterol) examined as a secondary aim, as the physical exercise demands of soldiers in the Army may confound expected gene and adversity predictors of obesity. Discovery and replication samples will be constructed separately for the analyses predicting smoking and hazardous drinking to maximize power, with over 2,000 cases and 5,000 controls expected to be available for the discovery datasets, and over 2,000 cases and 5,000 controls available for the replication analyses for both traits. Subjects not exposed to tobacco or alcohol, approximately 6% of the Army STARRS cohort, will be excluded from these primary analyses. Discovery (N=8,000) and replication (8,000) samples for the secondary dimensional obesity indices will use the entire Army STARRS cohort (N=16,000). Gene-Environment (G-E) correlations will be explored first in the dataset, and then the most appropriate statistical approach selected for the gene-environment-wide interaction analysis (GEWIS), given empirical hierarchical Bayes approaches have been found to be more optimal in the presence of numerous G-E correlations, and the two-step approach delineated by Murcray and colleagues preferable if there are no or only a few weak G-E correlations. As methylation and gene expression data are available for a subset of Army STARRS participants (N=400), exploratory analyses will also be conducted focusing on genes identified in the gene-environment-wide interaction analyses. The investigators on this team will have the capacity to follow-up on R21 research findings in additional diverse populations (not to be funded with this R21), and the availability of strong well-validated animal models to study the effects of stress can be used to support translational research efforts to confirm and further investigate GEWIS results.
Early adversity and stress later in life are common experiences that are associated with highly prevalent health risk behaviors (e.g., smoking, hazardous drinking, obesity), with each of these health risk behaviors associated with increased incidence of a broad range of medical and other problems. The goal of this project is to understand the genetic factors that interact with stress to confer risk for these problems.