Stranded with a blown-out rudder in the middle of Los Angeles' shipping lanes last summer, Trey Ideker drifted for three hours before he could repair his sailboat and get back on course. It was one of the few times someone might call him rudderless.
While at Massachusetts Institute of Technology, the computer science and engineering major had chosen a traditional route tinkering with machines. Then as a senior, he took freshman biology and discovered that what he had always considered a "soft science" was much more quantitative than he thought. "That's when I realized, hey, cells are little machines."
He was applying to graduate schools in his majors when a friend tipped him off to the ideal marriage of his original and burgeoning interests: Leroy Hood was recruiting computer scientists to his molecular biology program at University of Washington. Emboldened by "the realization that I wasn't giving up computer science as a field," Ideker joined the program in Seattle and has stayed the course.
Under Hood's direction, Ideker picked up classical biology and statistical techniques to complement his modeling skills. He applied both to look globally at galactose metabolism in yeast, by integrating DNA microarrays, quantitative proteomics, and protein interaction data to identify cellular responses to 20 perturbations he modeled in the pathway.
The engineer doesn't mind biology's deep waters. "We're not building cells to our specification yet," says Ideker, but current technologies provide a wealth of high-throughput information about gene expression waiting to be probed by specific hypotheses. "That's exactly what computers are designed to do."
Returning to Cambridge, Ideker joined the Whitehead Institute's coveted fellow program. There, he worked with Hood and others to introduce Cytoscape, an open-source software platform that allows scientists to create conceptual networks from databases of protein-protein, protein-DNA, and genetic interactions.
Since starting his lab at University of California, San Diego, Ideker has applied such tools to tackle everything from malaria to DNA damage. With his 20-member crew, he is also comparing species of virus, or cancer and normal cells, for example, based on the wiring of their internal networks. "He's got really terrific taste in problems," says Hood.
Ideker's hallmark is "knowing where the action is going to be," concurs Richard Karp at the University of California, Berkeley, Ideker's longtime advisor and collaborator. With rudder in place, he's bound to get there.
Title: Assistant professor of bioengineering, University of California, San Diego
1. T. Ideker et al., "Integrated genomic and proteomic analyses of a systematically perturbed metabolic network," Science, 292:929-34, 2001. (Cited in 584 papers) 2. P. Shannon et al., "Cytoscape: A software environment for integrated models of biomolecular interaction networks," Genome Res, 13:2498-504, 2003. (Cited in 105 papers).