A team of researchers at the Icahn School of Medicine at Mount Sinai have utilized an advanced artificial intelligence tool to pinpoint rare coding variants in 17 genes. These variants provide insight into the molecular basis of coronary artery disease (CAD), which is the leading cause of morbidity and mortality globally. The findings uncover genetic factors that impact heart disease and offer new possibilities for targeted treatments and personalized approaches to cardiovascular care.disease (CAD), which is the primary cause of illness and death globally.
The findings, outlined in the June 11 online edition of Nature Genetics, expose genetic elements that influence heart disease, paving the way for targeted treatments and personalized approaches to heart care.
The researchers utilized an in silico, or computer-derived, score for coronary artery disease (ISCAD) that comprehensively represents CAD, as explained in a previous publication by the team in The Lancet. The ISCAD score encompasses numerous clinical characteristics from the electronic health record,including important measurements, test results, prescribed drugs, symptoms, and medical conditions. In order to develop the scoring system, researchers utilized machine learning models to analyze the electronic health records of 604,914 people from the UK Biobank, All of Us Research Program, and BioMe Biobank in a thorough meta-analysis.
The effectiveness of the score was then examined for its correlation with rare and ultra-rare coding variations identified in the exome sequences of these individuals. Furthermore, the research team delved into further exploration of the identified genes to investigate their roles in causal CAD risk factors, clinical presentations of CAD, and their associations withThe traditional large-scale genome-wide association studies have revealed the status of CAD, among other factors.
“Our results shed light on the involvement of these 17 genes in coronary artery disease. Some of these genes are already known to impact the development of heart disease, while others have not previously been associated with it,” says Ron Do, PhD, senior author of the study and the Charles Bronfman Professor in Personalized Medicine at Icahn Mount Sinai. “Our research demonstrates how machine learning tools can uncover genetic insights that traditional methods might overlook when comparing cases and controls. This could lead to new ways of identifying biological mechanisms.”
Heart disease and gene targets for treatment may be affected by rare coding variants that occur in only a small percentage of individuals. These variants can have a significant impact on disease risk or susceptibility when present, making it essential to study them in order to understand the genetic basis of diseases and inform therapeutic targets.
Over the last decade, traditional methods relying on diagnosed cases and controls have faced challenges in identifying rare coding variants associated with CAD. The limitations of diagnostic codes in capturing the complexity of CAD prompted researchers to explore new methods.
“Our previous Lancet study demonstrated that a machine learning model trained using electronic health records has the ability to generate a virtual score for coronary artery disease, capturing the disease in all its forms,” states lead researcher Ben Omega Petrazzini, BS, Associate Bioinformatician in Dr. Do’s lab at Icahn Mount Sinai. “Building on these results, we theorized that the in-silico score for CAD could uncover new rare coding variations associated with CAD by providing a more comprehensive understanding of the disease.”
Moving forward, the team plans to further explore the role of the identified genes in the biology of CAD.The researchers are looking into how machine learning could be used in genetic research for complex diseases. This is part of their work to better understand how diseases work, find new treatments, and help patients.
