Staff members involved in high-performance computing at the US Department of Energy’s (DOE’s) Oak Ridge National Laboratory (ORNL) stand at the intersection of many scientific disciplines and try to create computational innovations that impact as many of those disciplines as possible.
Over the last several years, an ORNL team led by Sreenivas Rangan Sukumar, former group leader for the Oak Ridge Leadership Computing Facility’s Advanced Data and Workflow Group, has been developing a computational tool for connecting the dots between disparate medical discoveries recorded in the literature.
Despite his lack of medical background, Sukumar spearheaded the development, along with collaborators at the US National Library of Medicine (Dr. Thomas Rindfleisch) and University of Maryland School of Medicine (Dr. Eliot Siegel), of the Oak Ridge Graph Analytics for Medical Innovation (ORiGAMI)—a tool that can help doctors hypothesize and discover connections between patient symptoms, diseases, drug interactions and side effects, and more.
Through its recent success in identifying historical figures’ causes of death, ORiGAMI has raised its profile nationally. It was recently made open source by DOE and is a finalist for the R&D 100 Awards—a prestigious international award bestowed by R&D Magazine for the year’s outstanding innovations. For more than 50 years, the R&D 100 Awards have recognized excellence in innovation, earning the name the “Oscars of Invention.” This year’s event will take place November 2–4 in Washington, DC.
“Humans’ limited bandwidth constrains the ability to reason with the vast amounts of available medical information,” Sukumar said. “By design, ORiGAMI reasons with the knowledge of every published medical paper every time a clinical researcher uses the tool. By allowing computers to do what they do best, doctors can do better at answering health-related questions.”
ORiGAMI channels the vast knowledge stored in the National Library of Medicine’s MEDLINE database. MEDLINE aggregates information from more than 5,600 scientific journals pertaining to heath, medicine, or other life sciences.
This database has long been doctors’ best resource for gathering information about drug interactions and side effects, but finding disparate connections between symptoms and illnesses or drugs, physiology, and side effects can be an extremely labor-intensive process given anywhere from 2,000 to 4,000 new citations are entered into MEDLINE every day.
ORiGAMI doesn’t just sift through mounds of MEDLINE data and pull relevant search terms, like Google tabulating your search results. It goes one further. ORiGAMI takes disparate data and maps connections and ranks, for instance, how deeply connected Ebola and an antimalarial drug may be or how much more likely it is that people who have high blood pressure will experience a certain side effect of a drug.
“This technology embodies next-generation search, where weakly correlated data and domain-specific context matter,” Sukumar said.
If you would like to try a demonstration of ORiGAMI, please click here. For more information on ORiGAMI’s open-source version, please visit https://github.com/ssrangan/origami.
Oak Ridge National Laboratory is supported by the US Department of Energy’s Office of Science. The single largest supporter of basic research in the physical sciences in the United States, the Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.