Researchers establish national center for biological simulations

The National Institutes of Health has awarded $19.9 million over five years to Russ Altman, associate professor of genetics, and Scott Delp, associate professor and chair of bioengineering, to establish and lead the National Center for Physics-Based Simulation of Biological Structures (SimBioS). The new center is charged with developing a simulation toolkit to enable scientists worldwide to model biological systems ranging from molecules to whole organisms.

SimBioS is one of four new national centers endeavoring to build the biomedical computing infrastructure urgently needed to speed progress in biomedical research. The centers will create innovative software programs and other tools allowing the biomedical community to integrate, analyze, model, simulate and share data on human health and disease. Establishment of the four centers this month was an important step in the National Institutes of Health Roadmap, a series of far-reaching initiatives designed to transform the nation's medical research capabilities and speed the transfer of research discoveries from the benchtop to the bedside.

"There has been over the last couple of decades a lot of progress in simulating biological structures in order to understand how they work and how they function," said Altman, a practicing physician and bioinformatics expert who holds departmental appointments in Genetics, Medicine and Bioengineering, as well as a courtesy appointment in Computer Science. "We would like to bring this capability to all biologists in a routine way."

The grant, which holds the possibility for renewal for another five years, aims to build an easy-to-use software package allowing high-quality physical modeling. Altman's vision? "A biologist working on a problem at the molecular, cellular or organismal level may have questions about how the physics of their system affects its function. Using our software, they will, without having to establish new collaborations or going back to school, have a well-supported tool to do the initial modeling. We think this is within grasp."

The SimBioS Center will be located in the Clark Center in shared spaces that the biocomputation faculty has allocated for just this sort of collaboration.

"One of the challenging elements of the project is our goal to simulate a range of living systems from individual atoms to entire organisms," said Delp, who is a bioengineer and expert in biomechanical modeling and simulation. "If you look at the physics of many different biological structures, the fundamental equations that describe these systems have a lot in common. We're going to create a package that can simulate everything from proteins folding to human movement."

Mathematical and computational models provide the framework for understanding complexity in life-a requirement for further progress in biology and medicine. "The group at Stanford has a wide range of expertise, but a particular strength in physics-based simulation of biologic structures," Delp said. "Russ brought together the faculty members from Computer Science, Structural Biology, Biochemistry, Genetics, Mechanical Engineering, Bioengineering and many other departments to define a vision for how Stanford could provide the computing infrastructure for the nation, and the world, to manage biological complexity."

Many faculty played leadership roles during the three years of planning that led to the grant proposal. They include computer science core researchers Leo Guibas, Oussama Khatib, Adrian Lew, Ron Fedkiw, Pat Hanrahan and Jean-Claude Latombe and biomedical computation core researchers Michael Levitt, Vijay Pande, Charles Taylor and Sandy Napel.

Those leaders also include faculty working on four "driving biological problems"-physics-based challenges that are archetypes of common problems in a field, cover a range of scales and have a lot of experimental data from the real world to provide the constraints necessary to drive and validate models and simulations. The problems under investigation are all currently existing peer-reviewed projects: RNA folding, which has implications for rheumatological diseases (principal investigator Dan Herschlag); myosin dynamics, which is important for understanding myopathies and generation of motive force throughout organ systems (Jim Spudich); neuromuscular dynamics, which must be understood to better treat movement disorders resulting from cerebral palsy, stroke and Parkinson's disease (Scott Delp); and cardiovascular dynamics, which has important implications for coronary artery and peripheral vascular disease (Chris Zarins).

Addressing these problems "makes sure we make contributions soon to the science," Altman said. "We have scientists in those domains who've signed up to be early users of our software and to give us feedback as to whether it's working or not working and what it needs to do."

The center also will support creation of new courses in the Bioengineering Department and a newsletter about biomedical computation geared for the public. It will provide on-site training to users of the simulation toolkit, as well as streaming online instruction through the Stanford Center for Professional Development.