A young Asian boy holding a red heart in both hands to symbolize heart shape and links with congenital heart defect (CHD)
Credit: krisanapong detraphiphat/Getty Images.

Researchers at the Icahn School of Medicine at Mount Sinai have made significant strides in understanding the genes behind of coronary artery disease (CAD) through the use of an advanced artificial intelligence tool.

Published in Nature Genetics, their study identified rare coding variants in 17 genes that provide new insights into the molecular basis of CAD, the leading cause of morbidity and mortality worldwide.

The research team, led by Ron Do, PhD, and Ben Omega Petrazzini used an in silico score for coronary artery disease (ISCAD) developed from a previous study. This score incorporates a vast array of clinical features from electronic health records, including vital signs, lab results, medications, symptoms, and diagnoses.

By training machine learning models on data from 604,914 individuals across the UK Biobank, All of Us Research Program, and BioMe Biobank, the team created a comprehensive meta-analysis that accurately represents CAD.

Using this AI-driven ISCAD score, the researchers conducted a rare variant association study with exome sequencing data, uncovering rare and ultra-rare coding variants in the genomes of the study participants. These variants, present in only a small percentage of individuals, can have significant impacts on disease risk or susceptibility. The study not only highlighted previously known genes associated with heart disease but also discovered new genetic factors that had never been linked to CAD before.

“Our findings help us understand how these 17 genes are involved in coronary artery disease. Some of these genes are already known to influence heart disease development, while others have never been linked to it before,” explained Ron Do, senior study author and the Charles Bronfman Professor in Personalized Medicine at Icahn Mount Sinai.

“Our study shows how machine learning tools can uncover genetic insights that traditional methods might miss when comparing cases and controls. This could lead to new ways to identify biological mechanisms of heart disease or gene targets for treatment.”

The study’s innovative approach addresses the limitations of traditional methods, which often rely on diagnosed cases and controls and can miss the complexity of CAD. The researchers’ use of machine learning models to analyze electronic health records offered a more holistic view of the disease, revealing novel rare coding variants related to CAD.

“Based on our previous work, we hypothesized that the in-silico score for CAD could reveal novel rare coding variants by providing a more comprehensive picture of the disease,” said lead author Ben Omega Petrazzini. “This approach has proven successful, highlighting the potential of AI tools in uncovering genetic factors that influence complex diseases.”

The team plans to further investigate the roles of the identified genes in CAD biology and explore the applications of machine learning in studying other complex diseases. This ongoing research aims to advance the understanding of disease mechanisms, discover new therapeutic targets, and ultimately improve patient outcomes.

Also of Interest