COVID-19 Update: The university remains under a state of emergency. Learn more about guidelines and requirements in place for return to our campuses at Safe and Healthy Buckeyes.
Drug repurposing knowledge graph could help find COVID-19 treatments
The fight against COVID-19 has provided a host of challenges to experts in varying fields to find ways that improve or redesign current systems in a way better suited for international pandemic. Among these systems is drug discovery, by which new medications and remedies are discovered via ingredient identification or serendipity. However, thanks to the power of big data and the work of a team including Computer Science and Engineering Assistant Professor Xia Ning, an entirely new model for drug discovery has been developed for public use.
The team of scientists and engineers, including representatives from University of Minnesota, Hunan University, and Amazon’s AWS AI laboratories, have developed what is called the Drug Repurposing Knowledge Graph (DRKG). Drug repurposing is a drug discovery paradigm that uses existing drugs for new therapeutic indications. It has the advantages of significantly reducing the time and cost compared to de novo drug discovery, and due to its efficient design, may be of significant use in fighting the COVID-19 pandemic.
DRKG itself is a comprehensive biological knowledge graph that relates human genes, compounds, biological processes, drug side effects, diseases and symptoms. It curates and normalizes information from six publicly available databases and data that were collected from recent publications related to COVID-19. The graph shows how each component is linked and reveals potential interactions. The goal is to provide other researchers with a list of drugs that will narrow their search, reduce time and save money.
Along with the graph, Ning and team also developed a set of machine learning tools that can be used to prioritize drugs for repurposing studies. The tools use the state-of-the-art deep graph learning methods (DGL-KE) to compute embeddings of DRKG entities and relations, and use these embeddings to predict how likely a drug can treat a disease or how likely a drug can bind to a protein associated with the disease. When tested against the human proteins associated with COVID-19, these tools identified many of the COVID-19 drug candidates that are currently in clinical trials.
Ning and her collaborators have made DRKG publicly available on github along with the set of machine learning tools and pre-computed embeddings. This free infrastructure will ultimately facilitate researchers to conduct computational drug repurposing more efficiently and effectively for COVID-19 and for other diseases.
Ning also co-directs the Translational Data Analytics Institute's Computational Health & Life Sciences community of practice.
from the Translational Data Analytics Institute