By Mark D. Uehling
October 15, 2002 | The cell is a silent, windowless factory -- a chemical black box. Yes, thanks to genomics, scientists do have a sheaf of engineering blueprints of everything that can be built inside it. And thanks to proteomics, researchers know some of the molecules that eventually emerge from the factory.
But scientists still can’t calculate raw numbers of compounds or the rates at which all of a cell’s chemical components are transformed into other components. “This problem -- predicting how the system is going to behave -- is one of the important agenda items for 21st century biology,” says Roger Brent, president and research director of the Molecular Sciences Institute (MSI. “It’s one of a few massively important problems.”
However, this former professor of genetics at Harvard Medical School now says he may be able to crack open the black box -- at least for cells of the lowly baker’s yeast. In simple terms, he’s talking systems biology at a molecular level. Brent and the MSI nabbed a $15.5 million grant over five years in August from the National Human Genome Research Institute (NHGRI) of the National Institutes of Health. The grant backs MSI's Alpha Project, which aims to produce an interactive model of the pheromone signal transduction pathway in yeast.
Taking Apart a GPCR
“Yeast pathways,” of course, sound like a blind alley; they aren’t. The pheremone pathway in question happens to be an example of a G-coupled protein receptor (GPCR). GPCRs are exhaustively studied molecular docking mechanisms. They are targeted by at least 50 percent of all drugs on the market, from allergy medicines to heart pills. Examples of GPCR-based drugs include Zantac, Allegra, BuSpar, and Imitrex.
The nonprofit, independent status of the Berkeley, Calif.-based MSI means that any deeper insights or software algorithms it develops about GPCRs will go into the public domain. Brent calls this strategy “open source biology.” It’s a contrast to other ongoing systems biology efforts at companies like Entelos, Physiome Sciences, and Gene Network Sciences. “There will be tools flowing from this,” Brent vows. “There already has been a trickle.”
The bad news is that any computer model of yeast pheromones will be extremely difficult to create. “The majority of research that NIH supports does not have this level of risk,” says Jeff Schloss, program director, technology development coordination at the NHGRI.
Schloss points out that MSI is one of the NHGRI’s five Centers of Excellence in Genomic Science, and all five are on the frontier of knowledge. “To try to really push the boundaries of what one can do in genomics -- that’s the purpose of this program,” Schloss says. “If you’re really pushing the boundaries, then you’re going to be doing things that could fail.”
Brent himself is happy to concede the Alpha Project faces at least one big technical challenge: Scientists can’t measure what any individual cell produces, let alone describe how it all fits together. “We’re pushing beyond present technology and attempting to develop new methods,” Brent says. “We have to bring into being new ways of measuring what we need to measure.”
If that problem -- and serious computational issues -- can be resolved, an Alpha application or set of algorithms could predict what a yeast cell would do in response to a perturbation from its environment or a drug. Brent can imagine a scientific meeting of yeast experts and the following scenario: “Members of the audience could send in boundary conditions on cards, saying ‘What do you think will happen if we reduce the concentration of this molecular species by a factor of five?’ We would go to our oracle, to the model, and it would spit out the answer,” he says. MSI researchers would then go into the lab and do the experiments, and the answer would be identical, Brent says.
To create a digital yeast oracle, Brent will be leaning heavily on MSI’s elite cross-disciplinary team. “If this can be attacked fruitfully in this era, it will be because of the multidisciplinary effort,” says Brent, citing a team than includes a physicist who helped find the top quark and a mathematician with a deep background in geometry.
The promising project and its team prompted Craig Venter to issue a rare public plug. “Unique, interdisciplinary approaches to large-scale scientific research like the Alpha Project will enable us to better understand human biology and cellular function,” said Venter, president of The Center for the Advancement of Genomics. “It is encouraging to see that the innovative ideas of Dr. Brent and others at the Molecular Sciences Institute are being recognized and awarded significant federal funding to further their crucial research.”
Brent will also have help from outside MSI. Collaborators include Caltech, MIT, the University of California-Berkeley, and Pacific Northwest National Laboratory. If things get really dicey, Brent has a few other pals who also stand ready to help-- friends with nice, big supercomputers.