Research Project
10389028
Project Summary Frequent alcohol use can lead to alcohol use disorder, which accounts for three million deaths and over 133 million life years lost to disability and death worldwide per year. Alcohol use outcomes are under genetic influence. Genome-wide association represents the state-of-the-science statistical methodology for identifying genes associated with alcohol use outcomes. However, contemporary GWAS methods typically do not account for variability in genetic effects throughout development. For example, when analyzing longitudinal data, most contemporary GWAS are developmentally agnostic and average across timepoints to construct a phenotype that disregards developmental variability in genetic effects. GWAS summary statistics are used as the starting point for phenotype prediction via polygenic risk scores (PRS), which aggregate measured genetic effects on a phenotype into a score that indexes the statistical association between SNPs and the phenotype. Models of genetic influences on alcohol use might be improved if they consider the phenotype as dynamic over time, rather than a fixed maximum or average. No previous studies have constructed PRS from developmentally- informative GWAS, that measure genetic effects that are specific to developmental stage and change over time. In this project, I will apply novel multivariate genomic methods to incorporate developmentally-informative phenotype data into GWAS, create PRS that reflect change over time in alcohol use and PRS that are specific to developmental stage, and validate findings in an independent sample. Longitudinal cohort studies targeted for gene-identification analyses include the Avon Longitudinal Study of Parents and Children (ALSPAC, n~10,000), the Collaborative Study on the Genetics of Alcoholism (COGA, n~2,000), and The National Longitudinal Study of Adolescent to Adult Health (Add Health, n~6,000). Genetic prediction analyses will be conducted in the Finnish Twin Cohort Study (Finn Twin, n~1,400). This project is consistent with NIAAA?s goal to take a lifespan approach to alcohol use and disorder, and provides an analytical approach for doing this for alcohol gene identification efforts. These novel methods will advance the field of behavior genetics beyond the study of aggregated longitudinal phenotypes and represents an important step towards the broader NIH goal of advancing precision medicine strategies for alcohol use outcomes.
