Machine Learning Model Identifies Potential Age-Defying Compounds for Future Complex Disease Treatment

Researchers at Scotland’s University of Edinburgh have developed a machine learning model that identifies chemical compounds that can eliminate senescent cells without harming healthy ones. Senescent cells occur in aging and disease, such as cancer, type 2 diabetes, osteoarthritis, and viral infections. Senolytics are drugs that target these cells and are gaining popularity, but only two have been shown to be effective in clinical studies. To avoid biasing the model, the dataset includes substances with both senolytic and non-senolytic characteristics. The algorithm was applied to a database of over 4,000 compounds, and 21 candidates were found. Three compounds, ginkgetin, periplocin, and oleandrin, were shown to eliminate senescent cells without affecting healthy cells. The findings suggest that the model’s use significantly reduces drug screening costs while identifying effective treatments for serious diseases.