May 16, 2024
The UNC Eshelman School of Pharmacy, in collaboration with the Renaissance Computing Institute (RENCI) at the University of North Carolina at Chapel Hill, is using data science and artificial intelligence (AI) to identify and validate clinical new uses of existing drugs for rare diseases, also known as drug repurposing.
The School and RENCI received a $3.2 million award from the Advanced Research Projects Agency for Health (ARPA-H), the first award in which Carolina is named by the agency. Being able to identify and validate new uses of drugs requires a critical initial step of hypothesis generation and computational validation, which will be the team’s focus in collaboration with five other research groups across the country.
The award will support the integration of dozens of biomedical databases that contain information about drugs, diseases and biological pathways that underly disease mechanisms. The most unique scientific objective is to conduct searches to see which drugs will work against which diseases (18 million possible connections). The hope is to find therapeutic options for 9,000 diseases that currently don’t have FDA-approved drugs.
“As a key scientific strategy, we employ a biomedical knowledge graph approach, specifically, ROBOKOP knowledge graph that leverages studies we have conducted at RENCI in the last six years as part of the National Center for Advancing Translational Sciences-funded biomedical data translator project,” said principal investigator (PI) Alex Tropsha, Ph.D., K.H. Lee Distinguished Professor in the Division of Chemical Biology and Medicinal Chemistry and chief domain scientist for molecular informatics at RENCI.
“We anticipate that our work in integrating knowledge into graphs, and making predictions based on those graphs will allow us to hypothesize new drug repurposing opportunities, with the potential to change lives,” said co-PI Chris Bizon, Ph.D., director of analytics and data science at RENCI.
The team will develop and use machine learning and AI tools to mine the integrated knowledge graph to identify the most probable novel drug-disease connections that can be validated experimentally, including clinical studies.
“The key expected impact is that our specific drug repurposing hypotheses will be validated in the clinic,” said Tropsha. “Of course, we also expect new computational approaches, tools and impactful publications but the practical outcome is by far the most important.”
The award was first announced at the White House Rare Disease Forum in February for the overarching $48.3 million project led by Every Cure, a non-profit in Philadelphia. The project is called MATRIX (Machine learning/AI-enabled Therapeutic Repurposing In eXtended uses). Tropsha and Bizon serve on the advisory board for Every Cure. One of the collaborating teams led by Melissa Haendel, Ph.D., just moved to the UNC School of Medicine from the University of Colorado and will continue to work with Tropsha and Bizon.