The NIH grant is keyed to genomics, while the NSF grant is aimed at physical structure, dynamics, and instrumentation, said Dan Rokhsar, a Berkeley physics professor and principal investigator on both grants. Though not formally linked, the programs financed by the two grants are expected to have considerable overlap.

More than 10 departments at the university and the Lawrence Berkeley National Laboratory — which is adjacent to the Berkeley campus and has a long history with genomics — are involved. LBNL’s Advanced Light Source and National Center for Electron Microscopy will provide the instrumentation infrastructure.

In turn, LBNL is connected to the Department of Energy’s Joint Genome Center, where Rokhsar is also the acting associate director for computational genomics. JGI, in Walnut Creek, about a half hour drive from Berkeley, amalgamates the sequencing efforts from three National Labs: Berkeley, Livermore, and Los Alamos. Berkeley’s new Neurosciences Institute and Health Sciences Center, associated with the medical school at UC San Francisco, are also tied in.

Given the 10-plus years already needed for advanced scientific training, how can tomorrow’s bioinformaticists be expected to shoulder still greater knowledge burdens?

Rather than approaching bioinformatics like a traditional degree course, Rokhsar explained that the programs will use team teaching, workshops, and retreats to immerse students in the scientific challenges of the moment.

“No individual can know everything” in a multidisciplinary world, he said. Instead, members of bioinformatics groups must learn to communicate despite the different jargons, notations, literatures, and worldviews of their respective backgrounds. “It’s not so important that a biologist learn a lot of physics or computer science, but they do have to be able to talk to people from those backgrounds.” Students are apt to be more nimble than faculty at hopping across the disciplinary boundaries, Rokhsar said.

Ehud Isacoff, a professor of biophysics at Berkeley and a principal investigator on the NSF grant, added that because computational genomics and structural biology do not yet interact closely enough, the two programs’ synergies will be valuable. “We still know virtually nothing about how proteins really operate in the cell,” he said.

“We’re beginning to get the picture that events are mediated by as many as 30 or 40 proteins stuck together in very particular ways,” Isacoff added.

The training programs will help generate not only the new methodologies to understand how the complexes form and communicate, but also “a whole new breed of scientists who can think about them as concrete objects, having specific structures and interactions and moving in real time,” Isacoff said.

According to Rokhsar, Berkeley is one of the few places in the world that has the educational platform, depth of research facilities, and scientific talent who are eager for cross-disciplinary interaction. Others are Stanford’s Clark Center, which brings together biomedical disciplines as diverse as radiology, mechanical engineering, and computer science, and the Bioinformatics Institute at UCLA.

—Potter Wickware