Abstract
Coronary artery disease (CAD) is the leading cause of death world-wide. It is a multifactorial genetic disease with an estimated heritability of 40%. Many genetic studies had been performed on CAD, but they explained only 10% of the total heritability, partially due to most of these studies were performed on Europeans only. Admixed populations, such as Hispanics, were understudied, despite their higher risks of CAD than Europeans. To explain additional heritability of CAD, 2023 individuals were ascertained from the Miami Cardiovascular Registry catheterization labs, which were representative of the South Florida population – a mixture of Hispanics, Europeans, and African Americans. The prevalence of CAD varies across different race-ethnicity populations. A causal risk allele with large distinct ancestral allele frequency differences can lead to local ancestry deviations at these causal loci. Therefore, I performed local ancestry inference using LAMP-LD/LAMP-ANC software, a Hidden Markov Model based method on our population. Summing local ancestry across the entire population, the study population was composed of 75% European ancestry, 20% African ancestry and 5% Native American ancestry. These results were corresponding to principal component method identified ancestry, which did not utilize any ancestry informative markers. Collateralization, a natural bypass process that protects against harmful CAD events, showed sex and ethnic differences in the presence of obstructive artery disease. In the study dataset, men had higher rate in collateralization than women (p= 0.000175). Interestingly, collateralization had distinct prevalence among race-ethnicity groups, with Hispanics being the highest (58%), followed by African Americans (50%) and Caucasians (47%). Due to different prevalence among race-ethnicity groups, I performed admixture mapping with inferred local and global ancestry and identified two genetic regions where local ancestry was associated with collateralization: Native American ancestry was associated with the presence of collaterals at a region on chromosome 17 (chr17:36120051–40782083, ß = 0.55, P-value = 0.0001); African ancestry associated with collaterals on chromosome 17 but at a different location (chr17:32796666–33440166, ß = 0.38, P-value = 0.0007). Gene-based test results showed that 8 genes within the Native American region and 5 genes within the African region were significantly associated with collateralization. Angiographically defined and clinically diagnosed CAD shared common and unique genetic variants. Previous studies on CAD mostly focused on clinically defined CAD, which included various sub-phenotypes such as chronical angina, myocardial infarction, etc. Instead, I used angiographic stenosis through catheterization imaging to define CAD and performed genome-wide scanning of genetic variants associated with it. I identified 6 SNPs and 7 genes that were significantly associated with both clinically and angiographically defined CAD. In my research, I studied genetic components for angiographic stenosis of CAD in a genetically diverse population. This phenotype allowed us to identify genetic components specific to the coronary angiographic stenosis rather than the complex disease. I also identified genetic variants and contributing factors for collateralization, a protective mechanism towards CAD. This may help us explain why some atherosclerosis patients had worse clinical outcomes than others, and whether the race-ethnicity prevalence differences in CAD could be partially attributable to collateralization. Clinically, genetic factors influencing CAD may lead to novel treatment for CAD and promote new opportunities for prevention of patients suffer from severe obstructive atherosclerosis. Last, studying this genetically diverse population allowed us to eliminate health disparities and therefore benefited all populations as a whole.