[K08] Partitioned polygenic risk scoring and machine learning-derived imaging phenotypes for discovery of novel mechanisms of cardiometabolic disease in people of South Asian ancestry
Ente: National Heart Lung and Blood Institute
Scadenza: 2031-07-31
Importo max: 169.348 EUR
Paese: US
Descrizione
Project Summary / Abstract:
Cardiovascular disease (CVD) is increasingly understood to be driven by metabolic dysfunction, as exemplified
by the global epidemic of obesity and type 2 diabetes (T2D) and the risk conferred for myocardial infarction and
heart failure. Individuals of South Asian ancestry face disproportionately high risks of myocardial infarction (2-4
fold higher) and T2D (4-to-6 fold higher) at younger ages and lower body mass indices. Thus, South Asians
represent a model highly enriched for unknown drivers of cardiometabolic disease and a highly efficient means
of therapeutic discovery – yet South Asians are dramatically understudied. Prior work has shown that South
Asians may have higher genomic risk for T2D and coronary artery disease through pathways relating to obesity,
lipodystrophy and hyper/hypo insulinemia. Dr. Bhattacharya proposes to use innovative computational
techniques to probe both the responsible genomic pathways, and the maladaptive adipose tissue storage
phenotypes that may exist in South Asians leading to increased risk of metabolic dysfunction and CVD. In Aim
1, Dr. Bhattacharya will use the Broad Institute’s Cardiovascular Disease Knowledge Portal to advance
partitioned polygenic risk scores by developing unbiased SNP-level and gene set-level scores to identify new
genomic subclusters of T2D and coronary artery disease. In Aim 2, Dr. Bhattacharya will apply deep-learning
methods to MRI/CT images for quantifying ectopic adipose distribution, and deriving data-driven clusters of
adiposity phenotypes (e.g. hepatic-predominant, visceral predominant, intramuscular predominant) and
discovering if these ML-derived phenotypes can predict myocardial infarction. In Aim 3, Dr. Bhattacharya will
evaluate whether distinct partitioned polygenic risk score clusters (e.g., obesity- vs. lipodystrophy-like) predict
enhanced responses to diabetes treatments like GLP1 receptor agonists in both biobanks (MGB Biobank, UK
Biobank) and the REWIND clinical trial. By integrating cutting-edge genomic analyses with deep-learning–based
imaging and pharmacogenomic studies, this work will improve risk prediction and inform more precise
interventions for high-risk populations, thus fulfilling the NHLBI’s mission to advance innovative strategies that
reduce cardiovascular morbidity globally. Furthermore, mechanistic and genomic insights in a high-risk cohort
hold great promise for the identification of targeted therapeutics in the age of genomically-informed medicines
for all individuals who suffer from metabolic and cardiovascular disease – increasing the potential reach of the
current work. Dr. Bhattacharya is a preventive cardiologist transitioning to an independent physician-scientist.
His career development plan includes intensive training in computational genomics, machine-learning–driven
image analysis, and pharmacogenomics, supported by a mentoring team of leaders at Massachusetts General
Hospital and the Broad Institute. This environm
Istituzione: MASSACHUSETTS GENERAL HOSPITAL
PI: Romit Bhattacharya
Progetto: 1K08HL183778-01A1
Settori: National Heart Lung and Blood Institute
Vai al bando originale
Registrati gratis su Bandolo per trovare bandi compatibili con la tua azienda.