The Human Genome Project may be officially wrapped up, but that doesn’t mean that the days of federally funded big biology are over. In fact, a systems biology funding initiative at NIH’s National Institute of General Medical Sciences may do just as much — if not more — to advance computational biology as the National Human Genome Research Institute has.
Last week, NIGMS awarded the latest in a series of multimillion-dollar, five-year grants to create so-called Centers of Excellence in Complex Biomedical Systems Research — multi-disciplinary hotbeds where biologists, computer scientists, physicists, engineers, and mathematicians can bring the tools of quantitative science to bear on biological research. The Massachusetts Institute of Technology’s Computational and Systems Biology Initiative (CSBi) received around $16 million to create a Center of Excellence in Cell Decision Processes (CDP), and Harvard University’s Bauer Center for Genomics Research received around $15 million to study how “functional modules” of genes and proteins work together to control cellular systems.
A year ago, NIGMS committed around $25.5 million over five years to support similar centers at the University of Washington and Case Western Reserve University. The MIT and Harvard grants are the largest awarded so far in the Centers of Excellence program, which plans to award an additional $6 million in fiscal year 2004 for two or three more centers, with up to $2 million per center annually for four additional years (see sidebar, p. 7).
“Biologists come from a very different, and much less quantitative background than do the physical scientists,” said John Whitmarsh, a program director at NIGMS’ Center for Bioinformatics and Computational Biology. “So my interest, and that of many people here at NIGMS, is to enhance quantitative biology — simply measuring things with numbers, and then recognizing that biological systems are much more complex than anything we work with; that other formalisms need to be brought into the field in order to make sense of it.”
Whitmarsh, who has a background in physics, likened the state of biology today to that of physics in the second half of the twentieth century. “After World War II, there weren’t many experimental problems that could be solved by individuals, so multidisciplinary teams emerged — you started to see the number of authors on papers going from three to five to 20 to 50 to 100. Biology is undergoing that same change now,” he said. By backing multidisciplinary research centers for quantitative biology, he said, NIGMS is building the infrastructure to usher in that change.
It wasn’t always that way, however. As the “basic research” arm of NIH, NIGMS supports early-stage technology that many other NIH institutes wouldn’t touch, but systems biology was considered a bit too speculative by the funding agency until very recently. Researchers looking for funding to support multidisciplinary projects had to turn to less risk-averse funding bodies, like the Defense Advanced Research Project Agency, to support their work.
The MIT Cell Decision Processes program, for example, has its roots in DARPA’s “Bio:Info:Micro” program, which began in 2000. The project, which CSBi co-chair Bruce Tidor acknowledged was “blue-sky research” at the time, set out to answer the question of “how a cell decides to live or die” using a combination of experimental and computational methods. “The same sort of work at such early stages would not be funded by NIH,” Tidor said. Three years later, with some results in hand and a clear sense of where the project is headed, Tidor and his colleagues found an eager supporter in NIGMS.
Tidor said that the new funding would help the project expand its staff and range of technologies. Currently, 13 MIT faculty members are involved in the project, which is studying the response of a human cell line to growth factors like insulin and epidermal growth factor, as well as to death-inducing factors like tumor necrosis factor. Tidor and four other MIT faculty members are developing computational methods to analyze data collected from new, microfluidic technologies developed as part of the project, and are “tweaking out the underlying model from the experiments.”
Like other types of genomic and proteomic data, the information Tidor’s team is gathering is “noisy and non-linear,” but it offers the additional complexity of “direct and indirect effects, meaning that some things directly turn something on, but other things turn something on through a third agent. And trying to distinguish those possibilities is extremely important for building and understanding the underlying network,” he said. The CDP team has refined a “cyclic process of experiment, computation, experiment, computation” to build hypothetical biological networks and test them experimentally, Tidor said.
Tidor said he plans to use the NIGMS funding to expand the project’s proteomics capabilities and RNAi technologies, and bring the prototype microfabrication technology into production mode. From an informatics perspective, he said an important goal is to identify the most critical information for building network models from experimental data. “If we knew what data to collect, it might actually be much easier, but right now we don’t really know, so we’re collecting all kinds of data,” he said. “It could be in the end that we could throw out a lot of it and still get the right answer. But we still don’t know what data it is that we should be ignoring.”
The NIGMS award marks the first large grant that CSBi has received since it was founded in January, and makes up about half of the $30 million to $35 million the center has raised so far, said Brigitta Tadmor, executive director at CSBi. The center also relies on financial support from industrial partners “who want to be here just to see how this is all going to play out,” she said. “The whole concept is so new that industry is interested in seeing how MIT approaches it” before investing the resources in setting similar in-house projects, she added. So far, CSBi faculty members have partnered with IBM, Amgen, AstraZeneca, and others on different research projects.
While research institutes like CSBi and the Bauer Center already have multidisciplinary teams on hand to attract industry partners and federal funding, Whitmarsh stressed that the NIGMS complex biomedical systems initiative doesn’t intend to favor the large centers over smaller players. “That’s an issue that we’re very sensitive to, and it’s discussed a lot,” he said, adding that a number of NIGMS programs are geared toward individual investigators within academia as well as industry. “The plan is that we get some mathematicians into the system, they start training, and we broaden the pool of those people doing biomedical research,” he said.
While the agency may be more open to systems biology than it was a few years ago, “It’s too young to judge its contribution to understanding biomedical problems, and will be for years to come,” Whitmarsh said. “I think there is a sea change going on, and whether it’s successful or not really depends on the success of the programs we’re funding like this, as exemplars of what others can do. I’m committed to this, but that doesn’t mean that everybody at NIH or NIGMS is as convinced of the efficacy of this approach.”