SRA

Identification of causal variants and mechanisms underlying complex disease traits in humans requires strategies to locate and fine-map functional regulatory variants in disease-relevant cell types. To discover functional regulatory variants in primary aortic endothelial cells (ECs) from humans, we collected genetic, transcriptomic, and four epigenomic phenotypes in a population of up to 150 human donors representing individuals of both sexes and three major ancestries. We found thousands of EC eQTLs that were not present in GTEx at all ranges of effect sizes, indicating novel functional variants not observable from tissue data. We performed several epigenetic assays across 53 donors' EC to enable molecular QTL mapping, which included chromatin accessibility, histone modification, and transcription factor binding of proteins ERG and NF-kB/p65 in two activation states. We discovered over 3000 regulatory elements where cis-regulatory variants associated to significant differences in epigenetic marks. Co-localization with our eQTLs, TF motif mutations, and 3D confirmation capture data in ECs enabled our discovery of hundreds of high-confidence functional regulatory elements that perturb endothelial molecular phenotypes. Furthermore, this set of variants is enriched at GWAS loci for Coronary Artery disease and other disease traits, with some SNPs demonstrating pleiotropic effects. Together, our dataset and analytical approach provide the genetics and vascular biology communities with specific variants affecting EC biology and serves as a proof-of-principle for how to detect functional regulatory variants in human cells with relevance to complex traits. Overall design: Examination of molecular traits of 157 Human Aortic Endothelial Cell Samples as follows: RNA-seq in 53 samples in two treatments (control, IL-1B treated), Microarray expression from 156 (control) and 157 samples (OxPAPC treated), H3K27ac ChIP-seq in 42 samples in two treatments (control, IL-1B treated), ATAC-seq in 44 (control) and 43 (IL-1B treated) samples, ERG ChIP-seq in 22 samples (untreated), p65 ChIP-seq in 36 samples (IL-1B treated).