Deciphering the genetic basis of complex traits is a central goal of human genetics and precision medicine. Current research in my group aims to develop computational approaches that fill knowledge gaps in two areas. First, the success of genome-wide associations (GWAS) has largely confined to populations of European descent; minority individuals are under-represented, and as a result, our understanding of disease etiology in minority populations lags behind. We have a long-standing commitment to develop and apply novel analytic frameworks, which aim to accelerate biomedical research in minority populations. Building upon our existing work, we will develop genetic risk assessment approaches for minority individuals that judiciously leverage both trans-ethnic and ethnic-specific information. We will also develop models for characterizing the genetic basis underlying population-level phenotypic differences. A second area, which we will investigate in parallel, is to use molecular phenotypes, in particular proteomics data, to elucidate the biological relationship between GWAS loci and disease outcomes. A fundamental limitation of GWAS is that it does not reveal the mechanisms through which DNA-level variation manifests into phenotypes; this is particularly problematic because a large fraction of GWAS SNPs falls into non-coding regions. The mapping of gene expression quantitative traits (eQTL) using RNA expression has provided a rich source of information regarding gene regulation. On the other hand, the genetic basis of post-transcriptional regulation and its impact on complex traits and diseases is poorly understood. We hypothesize that proteomics data allows us to gain understanding about post-transcriptional regulation, and protein abundance provides a heritable marker linking between RNA and phenotypes. Making use of proteomic data, such as those generated through GTEx, we propose novel analytic approaches for uncovering pQTLs, for using protein as intermediate phenotype in disease risk assessment, and for identifying candidate proteins that link GWAS loci and phenotype. Ultimately, we envision that the ensemble of methods we develop, by capitalizing on multi-ethnic cohorts and multi-omics data, will contribute to the implementation of individualized prevention and intervention strategies for people of all races and ethnicities.
The proposed research will provide computational methods for uncovering genetic determinants of traits and diseases relevant to minority individuals, and for interpretation of genome-wide association studies (GWAS) results. It has broad impact because the methods developed will make efficient use of two kinds of valuable resources: large-scale multi-ethnic GWAS data and rapidly expanding molecular phenotype (RNA, proteomics and other -omics) data. The methods are generally applicable to a broad range of traits and diseases, and will offer important insights to guide future design and implementation of precision health that benefits all individuals, regardless of race or ethnicity.
