AI Accelerates Discovery of Antiviral Compounds for Hand, Foot and Mouth Disease with Limited Data

Scientists at the Perelman School of Medicine at the University of Pennsylvania have developed a novel approach that integrates artificial intelligence (AI) with traditional laboratory methods to identify potential antiviral compounds against human enterovirus 71 (EV71), the primary cause of hand, foot, and mouth disease. This breakthrough, published in Cell Reports Physical Science, demonstrates the efficiency of AI in predicting antiviral drugs even when limited experimental data is available. Hand, foot, and mouth disease is predominantly a childhood illness, often causing mild symptoms like rashes and fevers. However, EV71 infections can escalate to severe neurological complications, especially in young children and immunocompromised individuals. Currently, no FDA-approved antiviral treatments specifically target EV71, making this research particularly significant. The team, led by Dr. Ivan de la Fuente and including postdoctoral researcher Dr. Angela Cesaro, began with a small set of 36 small molecules. They used these to train a machine learning model to recognize specific chemical features and shapes that effectively inhibit viral activity. The AI then scored each compound based on its potential to block EV71. From the initial panel, the model identified eight top candidates. When tested in cell experiments, five of these compounds successfully slowed the virus's replication, a success rate approximately ten times higher than traditional screening methods. "Traditional drug discovery is a costly and time-consuming process, often taking several months just to identify a few promising leads," de la Fuente explained. "Our AI-driven method collapses this timeline into days, making it a powerful tool, particularly in situations where time and resources are limited." The five successful compounds were further analyzed using computer simulations to understand how they interact with the virus. These simulations revealed that the compounds bind to specific sites on the viral surface, potentially preventing the virus from altering its shape and entering host cells. This insight could significantly aid future research in developing targeted antiviral therapies. "The findings provide a blueprint for rapid antiviral drug discovery," Cesaro noted. "This method can be adapted to other viruses, regardless of whether they are emerging, reemerging, or well-known pathogens like polio." The collaboration between the de la Fuente lab, Procter & Gamble, and Cornell University highlights the interdisciplinary nature of modern scientific research. By combining expertise in virology, AI, and pharmaceutical sciences, the team has demonstrated the potential of AI to revolutionize drug discovery processes. Dr. de la Fuente emphasized the broad applicability of their approach: "Whether we face a new respiratory pathogen, another enterovirus, or a resurgence of viruses like polio, our AI-driven strategy can quickly identify promising drug candidates, accelerating the pace of research and potentially saving lives." Industry insiders praise the innovative use of AI in this study, noting its potential to streamline and enhance antiviral drug discovery. The University of Pennsylvania, known for its cutting-edge medical research, continues to push the boundaries of technological integration in healthcare. Procter & Gamble and Cornell University’s contributions underscore the value of public-private partnerships in advancing scientific knowledge and practical applications. This research not only offers hope for combating EV71 but also sets a precedent for leveraging AI in the fight against a wide array of viral threats. The efficient and data-driven approach could lead to faster and more effective responses to future outbreaks, marking a significant step forward in pharmaceutical science.