Alexander Young, Anderson School of Management, University of California - Los Angeles (UCLA)
Inferring genetic causation using randomization of genetic material within families and modeling intergenerational inequalities
Multiple phenomena contribute to genotype-phenotype associations: causal effects of alleles carried by the individual on the individual (direct genetic effects); effects of alleles in relative(s) through the environment, called indirect genetic effects (IGEs); and confounding due to population stratification and assortative mating (AM). Confounding factors can lead to spurious inferences in downstream applications of genome-wide association studies (GWASs). Family-based GWAS — by using random genetic variation within families — removes confounding from estimates of direct genetic effects. However, current family-based GWASs are limited in power due to relatively small sample sizes of genotyped first-degree relatives. We describe our software package, snipar, that works within a theoretical framework that treats parental genotypes as missing data and imputes them based on Mendelian laws. We show results from our meta-analysis of family-based GWAS studies across 16 cohorts and 34 phenotypes. We end by examining mathematical models for joint genetic and cultural transmission of traits from parents to offspring.
More information about this speaker may be found at https://alextisyoung.github.io/