Dissecting the effect of genetic variants on atherosclerosis: integrating bulk and single-cell transcriptomics

scRNAseq Stanford biobank (n = 5 coronary plaques) t-SNE1 t-SNE2 Relative cell fractions in BiKE cohort Risk factors CAD-related SNPs in GWAS Correlation with cell fractions (cfQTLs) Top hit variant: eQTLs for ARNTL Correlation with cell markers GSEA and PPI analysis In vitro mechanistic experiments Treatments Symptoms Survival BiKE (n = 127 carotid plaques) Associations of cell fractions with In uence of genetics on plaque composition Link between ARNTL and atherosclerosis Microarray Deconvolution E ect of genetic variants on atherosclerosis Bioinformatic deconvolution was performed on bulk microarray data from carotid plaques of the Biobank of Karolinska Endarterectomies (BiKE, n = 127) using publicly available scRNAseq (single-cell RNA sequencing) data of coronary plaques from the Stanford biobank (n = 5).The obtained relative cell fractions were correlated with patients' clinical data, including risk factors (sex, smoking, diabetes), medication (statins), symptomatic status, and survival.Quantitative trait locus (QTL) analyses were performed to predict the effect of coronary artery disease (CAD)-associated SNPs (single nucleotide polymorphisms) on different cell fractions (cfQTLs) and gene expression (eQTLs) in plaques.Several variants were mapped upstream of the Aryl Hydrocarbon Receptor Nuclear Translocator Like gene (ARNTL), whose expression in plaques negatively correlated with vascular smooth muscle cell (VSMC) markers.In vitro experiments suggest that ARNTL modulates VSMC phenotype and proliferation