Joe DeRisi

Later that night, in the break room just outside the lab, one team has taken over a couch next to stacks of empty pizza boxes, remnants of dinner. Laptops out, the students are discussing strange results from one of their experiments when a player from a competing team pops into the room. He lets them know about a paper that might explain what they saw.

This kind of interaction is what really excites the professors. Kortemme, who had most of these students in a class the previous quarter, says the winter challenge they did with DeRisi seemed to improve their classroom performance; they asked more questions and helped each other more than students in classes that hadn't been through the drill.

“Basically, they'd gotten comfortable enough that there are now no stupid questions, because everyone's stupid at some point,” she says.

And, everyone's expertise matters at some point. Charlie Kehoe, who has a master's degree in engineering from the Massachusetts Institute of Technology and spent two years at Amazon.com, Inc., before coming to UCSF, says that two quarters of biology coursework didn't give him the same level of “biological intuition” as his teammates who have biology degrees, but he's good at figuring out what kinds of problems are amenable to computational solutions. George Masologites, a biologist by training and Kehoe's teammate, wants to do experimental research, but he's already worked in a computational lab at UCSF because, he says, he wants to better understand how those labs approach questions. Just as important, he wants to forge relationships with scientists who have expertise in those areas—scientists with whom he might someday collaborate.

On Friday evening, when the four teams assemble to present their results, the teachers get a chance to learn just how collaborative their students have become. Each team found that one of the histone samples didn't behave as expected when analyzed with the protocol their professors had suggested.

“Something was going wrong, we weren't seeing any noticeable binding events,” says Masologites, describing the reaction that should have helped them analyze their sample—if it had worked. So they arranged to borrow time on a high-end spectrometry instrument that belongs to another professor, which allowed them to test whether the protein had unfolded from its normal shape. (It hadn't.) They also borrowed electrophoresis gels from a different lab to help determine the size of the proteins, as did some of the other groups.

After the presentations wrap up, the half dozen faculty members who have come out to watch gather in the hall to discuss the projects. DeRisi is particularly impressed with the creativity and resourcefulness of their approach, and several faculty say they plan to try some of the students' workarounds in their own labs. The actual results aren't bad, either. Kortemme got far more plausible design proposals than she expected. So many, in fact, that she and Narlikar need more time to analyze them before they pick the best of the best to build.

Photo: George Nikitin / AP, ©HHMI

	PAGE 2 OF 2	Back