Integrative genomic and functional genomic studies to connect variant to function for CAD GWAS loci

PROJECT SUMMARY Atherosclerotic coronary artery disease (CAD) remains the leading cause of death globally, despite effective therapies for many known risk factors. The majority of CAD loci identified by genome-wide association studies (GWAS) are not associated with traditional risk factors, providing opportunities to discover novel mechanisms and therapies. Encouragingly, many new therapies targeting causal genes and pathways inspired by GWASs have received approval or fared favorably in clinical trials. Despite the success, the bottleneck remains the lack of experimental studies systematically linking CAD variants to the causal genes and pathways at scale in a cell type-specific manner. To overcome these barriers and define tissue/cell type-specific contribution to the genetic predisposition of CAD risks, we leverage the STARNET eQTL data in nine CAD-relevant tissue/cell type from >1300 subjects and public eQTL data of immune and vascular cells. Aim 1 will apply advanced computational pipelines to nominate candidate causal CAD variants, cis-regulatory elements (cREs), and their target genes and related pathways at CAD loci in CAD-relevant tissue and cell types, providing tissue/cell type- specific mechanistic and therapeutic insights for CAD. Initial analysis supports that ~30% CAD risk loci are most strongly associated with eQTLs in macrophages or foamy macrophages compared with the other seven tissues in STARNET data. This is in line with the major role of circulating monocyte-derived macrophages in driving atherosclerosis. Leveraging the rich functional genomic datasets available for monocytes/macrophages, Aim 2 will apply mid-throughput functional assays, including single-cell CRISPR screening, massively parallel reporter assay, and arrayed knockout or overexpression cellular assays, to experimentally define the genetic contributions of monocytes/macrophages to CAD by connecting variants to genes and phenotypic roles, and build machine learning models predicting functional cREs and their target genes. Our study addresses unmet needs in the functional follow-up of CAD GWASs by performing integrative genomic analysis of CAD loci at an unprecedented scale, experimentally connecting variant to function in cell types critical for the genetic predisposition of CAD, building machine learning models for predicting functional cREs and their target genes for improved prioritization workflow, and providing a generalizable framework for extended discoveries in other CAD-relevant cell types. With the MPI’s expertise in statistical genetics, machine learning, macrophage biology, and functional genomics, our study has significant and broad impacts by providing (1) novel insights into tissue/cell type-specific contributions to the genetic predisposition of CAD that inform new biological mechanisms and therapeutic targets; (2) a catalog of phenotypically tested target genes highly likely to be causal for CAD, allowing immediate refocus to the most promising targets for accelerated translation, and (3) an innovative computational and experimental framework for systematic variant-to-function discoveries.