AI Predicts Genetic Mutations Driving Disease
- When genetic testing reveals a rare DNA mutation, doctors and patients are frequently left in the dark about what it actually means.
- The team set out to solve this problem using artificial intelligence (AI) and routine lab tests like cholesterol, blood counts, and kidney function.
- Traditional genetic studies frequently enough rely on a simple yes/no diagnosis to classify patients.
AI predicts Disease Risk from Genetic Mutations with High Accuracy
When genetic testing reveals a rare DNA mutation, doctors and patients are frequently left in the dark about what it actually means. Now,researchers at the Icahn School of Medicine at Mount Sinai have developed a powerful new way to determine whether a patient with a mutation is likely to actually develop disease,a concept known in genetics as penetrance.
The team set out to solve this problem using artificial intelligence (AI) and routine lab tests like cholesterol, blood counts, and kidney function. details of the findings were reported in the August 28 online issue of Science [1]. Their new method combines machine learning with electronic health records to offer a more accurate, data-driven view of genetic risk.
Traditional genetic studies frequently enough rely on a simple yes/no diagnosis to classify patients. But many diseases, like high blood pressure, diabetes, or cancer, don’t fit neatly into binary categories. The Mount sinai researchers trained AI models to quantify disease on a spectrum, offering more nuanced insight into how disease risk plays out in real life.
“We wanted to move beyond black-and-white answers that often leave patients and providers uncertain about what a genetic test result actually means,” says Ron Do, PhD, senior study author and the Charles Bronfman Professor in Personalized Medicine at the Icahn School of Medicine at Mount Sinai. “By using artificial intelligence and real-world lab data, such as cholesterol levels or blood counts that are already part of moast medical records, we can now better estimate how likely disease will develop in an individual with a specific genetic variant. It’s a much more nuanced, scalable, and accessible way to support precision medicine, especially when dealing with rare or ambiguous findings.”
Using more then 1 million electronic health records, the researchers built AI models for 10 common diseases. They then applied these models to people known to have rare genetic variants, generating a score between 0 and 1 that reflects the probability of developing the disease.
Understanding Genetic Penetrance and the Challenge of Rare Variants
Penetrance, in genetics, refers to the proportion of individuals with a specific genotype who actually express the associated phenotype (observable characteristic). A high-penetrance gene means that most people with the mutation will develop the disease, while a low-penetrance gene means many will not. Determining penetrance is crucial for understanding genetic risk, but it’s notably arduous with rare genetic variants.
The challenge lies in the limited number of individuals with these rare mutations. Traditional statistical methods require large sample sizes to accurately estimate penetrance. this leaves both doctors and patients uncertain about the clinical meaning of a positive genetic test result.As noted by the National Human Genome Research Institute, “most genetic variants discovered through genome-wide association studies have small effects, and many people carry variants of uncertain significance” [2].
How the AI Model Works: Combining EHR Data and machine Learning
The Mount Sinai team’s innovation lies in leveraging the wealth of data contained within Electronic Health Records (EHRs). Rather of relying solely on genetic information,the AI models incorporate routine clinical measurements – cholesterol levels,blood counts,kidney function tests,and more – to predict disease risk.
Here’s a breakdown of the process:
- Model Training: The researchers trained machine learning algorithms on a dataset of over 1 million EHRs from patients at Mount Sinai Health System. These models learned to associate specific combinations of lab values with the growth of 10 common diseases.
- Disease Selection: the 10 diseases included in the study were: atrial fibrillation, coronary artery disease, type 2 diabetes, chronic kidney disease, heart failure, high blood pressure, hyperlipidemia, osteoporosis, rheumatoid arthritis, and venous thromboembolism.
- Variant Request: The trained models were then applied to individuals with known rare genetic variants.
- Risk Scoring: