Imagine a future where a drug originally developed for one disease is quickly and safely repurposed to treat another — potentially saving years of research and millions of dollars. Thanks to the pioneering work of researchers at Oregon State University’s College of Engineering, that future is coming into focus. Stephen Ramsey, professor of computer science, and graduate student Frank Hodges are at the forefront of this drug repositioning effort. Their work is part of the Biomedical Data Translator Project, a multiinstitutional initiative funded by the National Center for Advancing Translational Sciences. The team includes 24 researchers from Oregon State, Penn State, the Institute of Systems Biology, the Broad Institute of MIT and Harvard, and the Université Grenoble Alpes in France. NCATS plans to fund the project with $12.8 million over five years. “The Translator Project aims to use the tools of artificial intelligence and distributed computing in order to enable biomedical researchers and clinicians to explore knowledge for the purpose of coming up with new therapeutic approaches for all kinds of diseases,” said Ramsey, who also holds an appointment in the Carlson College of Veterinary Medicine. The Translator Project integrates vast biomedical databases into a unified knowledge graph. This allows AI-powered reasoning agents to uncover hidden connections between drugs and diseases — connections that might otherwise go unnoticed. “Computers are great at sifting through large databases of facts to find connections between two different concepts, for example, between a gene and a disease or between a drug and a phenotype,” Ramsey said. This approach is particularly valuable for drug repositioning. Instead of starting from scratch, researchers can use Translator to ask complex questions, such as which existing drugs might impact a specific disease pathway. The system can reveal new therapeutic uses for drugs that are already approved and well-understood, potentially speeding up the path to clinical trials and patient care. The impact of this work is especially significant for rare diseases, which often lack dedicated treatments due to limited commercial incentives. By using AI to mine existing data, researchers can uncover overlooked therapeutic options. Ultimately, Ramsey and colleagues see AI-driven drug repositioning as a way to shorten the diagnostic journey, reduce unnecessary testing, and improve outcomes for patients — especially those with rare or hard-to-treat conditions. Their work exemplifies how engineering and computer science can come together to solve some of the most pressing challenges in healthcare, bringing hope to millions of patients worldwide. USING AI TO FIND NEW PURPOSES FOR EXISTING DRUGS WINTER 2026 OREGON STATE ENGINEERING 6 “COMPUTERS ARE GREAT AT SIFTING THROUGH LARGE DATABASES OF FACTS TO FIND CONNECTIONS BETWEEN TWO DIFFERENT CONCEPTS, FOR EXAMPLE, BETWEEN A GENE AND A DISEASE OR BETWEEN A DRUG AND A PHENOTYPE.”
RkJQdWJsaXNoZXIy MTcxMjMwNg==