Innovative Machine Learning Approaches Identify Potential Treatments for Emerging Zoonotic Viruses

Researchers from Southwest Research Institute (SwRI), the University of Texas at San Antonio (UTSA), and Texas Biomedical Research Institute (Texas Biomed) have made significant progress in discovering potential antiviral treatments for dangerous zoonotic pathogens. Utilizing SwRI's advanced Rhodium software, the team trained machine learning algorithms to virtually screen and analyze over 40 million chemical compounds, aiming to find candidates that can inhibit viruses capable of jumping from animals to humans, such as Nipah and Hendra viruses.

These viruses are endemic in certain regions and are known for causing highly lethal infections. The researchers focused on the protein structures of these viruses, including the measles virus, which shares a family connection with henipaviruses, to identify compounds with the potential to block viral activity. Their efforts resulted in the identification of 30 promising compounds that could serve as broad-spectrum antiviral agents.

Dr. Jonathan Bohmann of SwRI highlighted the potential of machine learning in rapidly pinpointing antiviral candidates, especially for viruses that are challenging to study under traditional biosafety constraints. This approach not only accelerates drug discovery but also optimizes resource utilization, as it permits virtual screening at a scale previously unattainable.

The significance of this research is underscored by the high mortality rates associated with Nipah and Hendra infections, which can reach up to 75%. By virtualizing the screening process, researchers can bypass the need for high-containment laboratories, thereby saving time and reducing costs. The collaborative effort demonstrates the power of multidisciplinary partnerships in addressing emerging infectious diseases and advancing the development of effective antiviral therapies.

This study also underscores the potential for such computational methods to contribute broadly to antiviral drug discovery, potentially leading to versatile treatments for multiple related viruses, including measles. As these viruses pose ongoing threats to global health, the integration of machine learning into virology research represents a promising frontier for pandemic preparedness and response.