A high-throughput, multimodal strategy to improve variant classification

About This Grant

PROJECT SUMMARY--------------------------------------------------------------------------------------------------------------------- The timely and accurate genetic diagnosis of Mendelian diseases enables interventions to prevent catastrophic outcomes such as progressive heart failure and metastatic cancer. These diagnoses depend on the ability to distinguish pathogenic variants from benign. However, the vast majority of variants that have been identified in disease genes are currently classified as variants of uncertain significance (VUS), which are not clinically actionable. This “VUS problem” represents a major barrier to the implementation of Genomic Medicine. This proposal aims to address the VUS problem by employing a novel, integrative approach to prospectively assess variant pathogenicity. This approach will leverage the enormous genotypic and phenotypic variation present in large, DNA-linked biobanks together with the results of high-throughput, in vitro functional assays with the goal of generating sufficient evidence to reclassify VUS as either pathogenic or benign, at scale. To capitalize on the PI’s expertise in computational electrophysiology and Cardiovascular (CV) Medicine, the proposed research will focus on CV disease-associated genes; however, the methods developed in this proposal will be designed to generalize to Mendelian disease in any organ systems. The overall aim of this proposal is to accelerate the implementation of Genomic Medicine through the prospective reclassification of disease-gene VUS. The first aim of this proposal is to develop and validate CV gene-specific phenotype risk scores (PheRS) in the large, DNA-linked biobanks All of Us, the UK Biobank, and Vanderbilt’s BioVU, which, collectively, will have whole genome sequences for over 1 million participants linked to phenotypic data. These PheRS will be developed to capture a spectrum of phenotypes detectable in the biobank records with the purpose of amplifying the often- faint phenotypic signal associated with rare pathogenic variants in population studies. The second aim of this proposal will generate evidence to reclassify CV disease-gene VUS in the biobanks by deploying the biobank- calibrated PheRS in conjunction with Multiplexed Assays of Variant Effects (MAVEs), high-throughput, in vitro assays which generate functional data for nearly all possible protein-coding variants in a gene of interest. This proposed research will be conducted at Vanderbilt University Medical Center (VUMC), under the guidance of an exemplary mentorship team and in parallel with a thoughtful career development plan. VUMC is a world leader in Genomic Medicine and Biomedical Informatics, and thus an ideal environment for the PI’s continued training. The career development plan has been tailored to develop further expertise in human genetics, biomedical informatics, and Genomic Medicine. Overall, the research, environment, mentorship, and training in this proposal will give the PI with the necessary experience to successfully transition into an independently- funded physician-scientist.