Systems Genetics of Human Aortic Endothelial Cells

Abstract

Human complex disease is driven by a multitude of factors including individual’s genetics and environmental stimuli. To deconvolute the effect of genetics on cell biology from the rest of the factors, my work focuses on the transcriptional and epigenetic consequences of inheriting disease associated loci. Essentially, by focusing on the immediate effects of genetic variation, we can apply mechanisms to variants that were previously functionally obscure. To identify functional mechanisms of genetic variation, we collected high-throughput sequencing data including gene expression, transcription factor binding, chromatin accessibility, and histone modifications as well as genotypes of 6.7 million SNPs in 53 human aortic endothelial cell samples from donors of varying genotypes, sexes, and ancestries. We used quantitative trait locus (QTL) mapping to single out over 700,000 SNPs that were associated with gene expression in our population. Of the ~700,000 expression QTLs discovered, 8,747 were associated with an epigenetic trait as well, such as chromatin accessibility, transcription factor binding, or histone modification, providing a mechanism for the association with target gene transcription. Additionally, the expression QTLs are enriched in high significance coronary artery disease associated loci meaning that we are identifying the possible pathways that drive increased risk for disease. Through QTL analyses, we discovered loci with multiple layers of evidence suggesting function within the human genome. To verify these candidates, we performed a massively parallel reporter assay called Self Transcribing Active Regulatory Region sequencing (STARR-seq) on 34,444 variants. Through STARR-seq, we were able to functionally validate 5,711 SNPs. Additionally, we found that chromatin accessibility and transcription factor binding were highly enriched within the validated set suggesting that these measures are vital when identifying functional variation from molecular data. In future work, we plan to investigate the role of genetic variation on gene regulatory networks in basal conditions as well as under conditions mimicking an inflammatory state similar to one seen in disease. We also are going to link individuals polygenic risk scores to their transcriptional networks and molecular QTL analyses to find patterns of mechanism that are acting to increase risk of developing coronary artery disease. Altogether, what is presented here is in-depth systems genetics on human aortic endothelial cells identifying disease relevant human genetic variation and the mechanisms through which they act on endothelial cell biology. The findings from this work have already benefited multiple labs and research projects and will continue to be an asset to the scientific community.