Print

NIH awards grant to launch major new biomedical computation center

Mark Musen

BY TRACIE WHITE

The National Institutes of Health announced Sept. 29 that it has awarded the School of Medicine a grant of $18.8 million to develop a National Center for Biomedical Ontology along with several other collaborating institutions. The goal of the center is to design and implement a new generation of computer systems that will enable researchers to share, compare and analyze data gathered from large biomedical experiments.

The center will be led by Mark Musen, MD, PhD, professor of medicine (medical informatics), whose Stanford research group created Protégé, considered to be the most widely used ontology-development software in the world.

"It felt like a call from Ed McMahon," said Musen, who was at a conference in Madrid in July when an NIH official contacted him. C. John Whitmarsh, PhD, acting director of the Center for Bioinformatics and Computational Biology, had called to inform him that his application was a serious contender for one of this year's three grants to develop a national biomedical computing center. "I was surprised we were as well-received as we were," said Musen. "We're probably a bit more cutting-edge than is typical."

For decades, scientists have developed computational tools to analyze data collected in laboratories and clinics with the aim of answering specific biological questions. Because many of these databases have been designed independent from each other, they are often incompatible, making it difficult, if not impossible, to use them together to reach conclusions. But recent advances in technology—in particular the growing popularity in the use of ontologies—could make it possible to integrate the data from divergent studies.

And that's where Musen's center comes into play.

"Simply put, ontologies are ways of structuring knowledge so that computers can use it," Musen explained. "Ontologies define in computer-understandable form the concepts—and the relationships among concepts—that are important in particular application areas. Ontologies have become critical for processing and integrating disparate data sources. The data integration provided by ontologies enables computers to draw inferences from diverse medical experiments, providing the kind of understanding that ultimately can help combat diseases."

Key to getting the grant funded was Musen's proposal to draw support from colleagues across North America and the United Kingdom, each with their own expertise. His application included the informatics group at the Berkeley Drosophila Genome Project—with its sustained support of the Gene Ontology Consortium and the Open Biomedical Ontologies library—as well as the Mayo Clinic, which is working to advance standards for the representation of clinical terminologies. Also involved are researchers from the University of Victoria in British Columbia and State University of New York-Buffalo.

In addition, three institutions will lead projects on the use of ontology services in addressing specific biomedical problems. Scientists at Cambridge University will explore how to use studies of mutations in fruit flies as models for understanding human diseases; researchers at the University of Oregon will explore the opportunities for analyzing relationships between gene function in zebrafish and humans, and researchers at the UC-San Francisco Medical Center will explore how to combine the results from multiple randomized trials in the area of HIV/AIDS.

The project with UCSF is a good example of how the center will try to make sense of difficult-to-reconcile results from randomized trials. According to Ida Sim, MD, PhD, associate professor of general internal medicine at UCSF, multiple trials have been conducted on reducing mother-to-child transmission of HIV and the benefits of treatment interruptions for AIDS patients on anti-retroviral drugs. But researchers have been unable to combine the results statistically because of their independent designs, she said.

Sim will use the information-management tools of the center to visualize, analyze and understand the valuable information from these trials to help advance HIV/AIDS care. It is hoped that research at the new center also will lead to new ontological software tools to resolve similar dilemmas in the future.

Musen and the NIH are now accepting applications from researchers at campuses across the country who wish to work with the center. If the agency accepts their proposals, these scientists can draw from NIH funding set aside specifically for this purpose. The NIH hopes to select more collaborative projects by early 2006.

Musen's center is the second National Center for Biomedical Computing established at Stanford since the NIH program began in 2004. Last year the agency funded the launch of the National Center for Physics-Based Simulation of Biological Structures, under Russ Altman, MD, PhD, professor of genetics, of medicine and, by courtesy, of computer science, and Scott Delp, PhD, professor of bioengineering and, by courtesy, of orthopedic surgery. That center is working to develop a simulation tool kit that will enable scientists to model and simulate biological systems from single atoms to entire organisms.