|
Abstract:
|
Many human traits and disorders are heritable, and recent large genome-wide association studies (GWAS) have identified individual genetic associations with a number of single-nucleotide polymorphisms (SNPs). However, GWAS have so far identified only a small fraction of the heritability of common diseases, so the ability to understand genetic mechanisms or to make meaningful predictions is still quite limited. Given the high number of complex, polygenic traits in humans (millions), and the relatively small number of genes (~20,000), it is certain that some genes affect multiple traits (i.e., are pleiotropic). Pleiotropy can be leveraged to improve power to discover new loci, to improve performance of polygenic risk scores, and to discover genetic pathways for complex traits and illnesses. In this talk our focus is on the association between schizophrenia and cardiovascular disease risk factors. Our statistical analysis is based on extending Efron's two-group mixture model for one phenotype to a 2^d-group mixture model for d>1 phenotypes. The method is illustrated with publicly available data on schizophrenia and cardiovascular disease risk factors.
|
