Researchers at the Perelman School of Medicine at the University of Pennsylvania have developed a novel artificial intelligence (AI) approach that identifies potential antiviral compounds against human enterovirus 71 (EV71)—the primary cause of hand, foot, and mouth disease—even when only limited experimental data is available.
The study, published in Cell Reports Physical Science, demonstrates that combining AI algorithms with traditional laboratory methods can reliably predict antiviral candidates. Using an initial set of 36 small molecules, the team trained a machine learning model to recognize specific chemical structures and features associated with antiviral activity.
This integrated approach significantly reduces the time and resources typically required for drug discovery. “We are collapsing what used to be months of trial-and-error into days,” said Dr. César de la Fuente, the study’s senior author.
The AI-driven method is particularly valuable in situations where time, budget, or data constraints limit traditional drug development processes. By efficiently identifying promising antiviral compounds, this technique holds promise for rapid response to emerging viral threats.
The research was conducted in collaboration with Procter & Gamble and Cornell University. The full study is available in Cell Reports Physical Science
