Gary H. Gibbons, M.D.
Senior Investigator, Metabolic, Cardiovascular and Inflammatory Disease Genomics Branch
Head, Cardiovascular Disease Section
Director, National Heart, Lung and Blood Institute
Scientific SummaryOne of the long-term goals of Dr. Gibbons' research program is to utilize new genomic technologies to integrate the areas of physiologic genomics, functional genomics and human molecular genetics in the field of vascular biology and medicine. Dr. Gibbons' translational research laboratory has a longstanding interest in elucidating the molecular mechanisms involved in vascular remodeling in health and disease. Our laboratory is interested in discovering novel mediators of vascular disease that may constitute candidate disease susceptibility variants. Under Dr. Gibbons' supervision, the CVDS is actively involved in collaborative projects designed to study the functional significance of epigenetic and genomic variation and the potential role of genetic variation in promoting the susceptibility to cardiovascular disease in both clinical and community-based settings. One of the ultimate goals of CVDS is to take a multi-level, multi-disciplinary approach to understanding and ameliorating racial and ethnic disparities in cardiovascular health.
One of the major objectives of the CVDS is to harness the rich African-American population in the Washington metropolitan area through leveraging the vast clinical and diagnostic resources of the NIH Clinical Center. The CVDS has recently established a new protocol to recruit over 3,200 African-Americans for clinical evaluation and research testing called GENE-FORECASTSM (GENomics, Environmental FactORs and the Social DEterminants of Cardiovascular Disease in African-Americans STudy). This resource will enable our team to test the working hypothesis that race-ancestry differences in the burden of cardiovascular disease (CVD) reflect the influence of a unique interplay between the distinct genomic variation characteristic of African-Americans and the 'exposome' of social determinants and environmental factors that influence the pathogenesis of CVD in African-Americans.
The objective of the CVDS is to develop a highly dense set of ancestry informative markers that transcends currently available panels. We are actively computing, imputing, and simulating genomic and phenotypic models developed by dimensionality reduction techniques to gain useful insights from big data (20 terabytes or larger). Currently, we are addressing this issue by actively recruiting participants into the GENE-FORECASTSM study and by integrating outside datasets. These include previously collected large CVD studies such as our MH-GRIDSM dataset, ClinSeq®, the Howard Family Health Study, and the 1000 Genomes Project (TGP).
Studying the involvement of rare and uncommon variants in complex trait etiology necessitates powerful and robust analysis methods. Bioinformatic tools are used as a first step to predict functionality. We are currently developing algorithms to model and predict a highly dense set of markers across the genome for differentiating sub-continental ancestry. Through utilization of various dimensionality reduction techniques, we aim to identify ancestry segments within each subject for inference of ancestral differences in novel putative functional variants and natural stratification for case/control comparison in variant enrichment analysis. We extend this family of approaches to include polymorphism frequency across populations (when available) as a metric for improved classification, as well as multi-way primate and mammalian alignments. Using these datasets, we intend to address fundamental questions linking biologically functional, ancestry-related DNA variants to CVD phenotypes.
The second objective of the CVDS is to examine the associations of social determinants, neighborhood characteristics, novel biomarkers (e.g., adiponectin), CVD phenotypes (e.g., coronary artery calcification, vascular function) or CVD risk factors (e.g., hypertension, obesity) in African-Americans. Data generated from surveys are used to assess the association between neighborhood characteristics, social determinant exposures, and self-reported CVD risk factors, after controlling for conventional covariates such as age and gender. Depending on the type of CVD phenotype, we perform various statistical analyses to examine the associations of social determinants with continuous CVD phenotypes (e.g., blood pressure, vascular function, Vitamin D levels) and categorical CVD risk morbidity (e.g., diabetes, obesity). In addition, we investigate associations of neighborhood characteristics with CVD risk factors. Models will include random components for neighborhoods to account for correlations between persons within neighborhoods. For assessing gene-environment interactions between minor allele frequency (MAF) and environmental exposures, logistic regression (binary outcomes) and linear regression (continuous outcome) will be used to control for covariates.