The goal of this proposal is to validate models of gene, environment, and developmental (GED) interplay for substance use disorders (SUDs). Much is already known about the developmental course and environmental risk for SUDs, and recent advances in genome wide association studies (GWAS) hold the promise for identifying novel risk genes for SUDs. Few studies, however, have integrated each of these factors into a programmatic line of research. This requires prospective cohort samples that have been genotyped and assessed on multiple occasions for SUD-related phenotypes and environmental risk factors (e.g., family, peers, school/work, stressful life events). Conducting new studies of this type with sufficient power to detect the small effects for individual genes and interactions with the environment will be extremely expensive and time consuming. We posit, however, that much of this knowledge can be obtained now by leveraging existing prospective cohort studies that have GWAS genotyping. We propose such a strategy to investigate GED interplay for SUDs. SUDs are excellent complex phenotypes to examine GED interplay, as they are common, heritable, and are associated with several environmental risk factors. SUDs are also ideal for examining development, as they cannot emerge prior to the discreet event of initiation, providing for a clear demarcation between a pre-morbid risk stage and an active risk stage. Finally, there are several existing samples that have been well characterized in terms of exposure to environmental risk and progression of SUDs from prior to initiation to severe and persistent problem use. We will conduct a series of analyses using the prospective twin and adoption studies of the Minnesota Center for Twin and Family Research (MCTFR; n=8405) and replication analyses in 7 longitudinal studies (combined n=7795; high-risk and population-based sampling design) to test and validate models of GED interplay. As the initial stage of GED interplay research will require winnowing down potential causal factors, Aim 1 is to develop polygenetic risk scores that aggregate the effects of multiple genetic markers to improve power to detect genetic effects.
Aim 2 is to model the developmental trajectories of SUD-related phenotypes using longitudinal mixed models, wherein age and environmental variables are used to account for individual differences in SUDs. Polygenetic risk scores are then added to the model to account for variation in the effects for age (G-D) and environmental (G-E) variables on SUDs.
Aim 3 is to replicate GED findings using the other samples for the purpose of meta-analyses, as this provides the greatest power to detect effects and the most reliable estimates of effect size.
Aim 4 is to expand our GED models to include other genomic data (exome and sequencing) as it becomes available. The richness of these data sets and our approach provides an especially cost efficient way to accelerate our understanding of the causal processes underlying SUDs, and we hope that cooperation across sites on this project will lead to continued collaborations that will further accelerate the pace of discovery.
Substance use disorders are complex phenotypes resulting from the interplay among genetic, environmental, and development risk factors, and create a substantial public health burden. We will validate an integrative model of to delineate GED interplay effects using existing prospective cohort samples that have been genotyped, which will improve our understanding of the etiology of substance use disorders and assist in relieving their public health consequences.
Showing the most recent 10 out of 17 publications
