In this project, we propose to use an integrative genomics approach to identify potentially functional regulatory variants in novel candidate genes for cardiovascular disease (CVD) risk. These objectively chosen candidate genes were obtained using large-scale transcriptional profiling of lymphocyte samples from 1,240 San Antonio Family Heart Study (SAFHS) participants. Target candidate loci were nominated based on the existence of significant correlations of quantitative gene expression levels with two major CVD risk factors, HDL-C levels and plasma triglyceride levels. Using our unique family-based resource of quantitative genome-wide transcriptional profiles, we will examine 50 novel candidate genes that exhibit both strong evidence for c/s-regulation of expression levels and significant correlations between expression levels and HDL-C or plasma triglyceride levels. Our prior linkage-based evidence for c/s-acting sequence variation can be exploited as a probabilistic causal anchor to maximize our chance for finding functional variation within proximal promoters. For each of these objectively chosen genes, we will (1) resequence approximately two kilobases of putative promoter region in 182 founder individuals to identify promoter variants; (2) genotype all detected promoter variation in the 1,240 SAFHS samples for which we have transcriptional profiles; (3) test whether promoter sequence variants are associated with gene expression levels of the appropriate candidate gene; (4) test for associations between promoter sequence variants and CVD-related phenotypes; (5) confirm observed associations in two independent samples, and (6) perform molecular functional analysis of the most likely regulatory variants The proposed integrative genomics research paradigm to be used in this project should increase the pace of discovery of the constituent genes of human quantitative trait loci (QTLs) influencing major risk factors for CVD risk. By focusing on genes whose transcripts show evidence for both c/s-regulatory variation and a strong relationship with the focal clinical phenotypes, we should maximize our probability for finding causal players in the CVD risk cascade.
Showing the most recent 10 out of 258 publications
