Integrative computational-experimental approaches for cardiovascular risk assessment

About This Grant

Project Summary/Abstract Cardiovascular disease (CVD) is the leading cause of morbidity and mortality in the US, with serum lipid levels, particularly LDL cholesterol (LDL-C), being a critical modifiable risk factor. Despite advances in genome-wide association studies (GWAS), which have identified over 400 loci associated with LDL-C levels, the specific genes and coding variants responsible for cholesterol regulation remain largely unknown. This knowledge gap hinders the development of individualized CVD risk assessments and targeted clinical interventions. Furthermore, many rare coding variants, including those in clinically important genes such as LDLR, APOB, and PCSK9, are too rare in population cohorts to generate reliable epidemiological risk estimates, resulting in many Variants of Uncertain Significance (VUS). While functional screening assays, including CRISPR-based technologies, have demonstrated great potential to assess the impact of rare variants, scaling these assays efficiently across the genome has been challenging. Functional screens provide valuable insights but are often limited by their in vitro nature, which may not fully capture the clinical impact of these variants. To address these challenges, we propose a comprehensive, integrative approach that combines large-scale population cohort data with high-throughput functional screening to improve our understanding of rare coding variants that influence lipid metabolism and CVD risk. Our approach leverages state-of-the-art base and prime editing platforms to functionally screen variants across 823 genes that have been shown to significantly impact lipid phenotypes. By integrating functional screening data with patient phenotypes from population biobanks, including the UK Biobank and AllofUs, we will directly test the relationship between functional data and clinical outcomes. This integration allows us to build robust gene-phenotype models, significantly improving the characterization of variants, particularly those with moderate or uncertain effects. For traits like LDL-C, our models will combine functional scores with patient- level factors, including polygenic risk, statin use, and lifestyle factors, to improve precision in risk prediction. In addition to generating functional data, we will prioritize variants for clinical reclassification, focusing on actionable disease genes present in research participants from large cohorts, including AllofUs and Mass General Brigham Biobank. By aligning our results with variant classification guidelines, we aim to directly improve patient care through the reclassification of VUS and enhanced cardiovascular risk stratification. Ultimately, this project will deliver a novel, integrative platform that accelerates the translation of genetic discoveries into clinical practice, improving outcomes for patients at risk of CVD.