PHILADELPHIA--When representatives of academia, industry and government addressed the future of genomics in a two-day seminar during the annual American Association for the Advancement of Science convention here last month, one point was made again and again: It will take a multidisciplinary approach to meet the challenge of functional genomics.
Sponsored by the Institute for Genomic Research (TIGR) and Science magazine, Genome Seminar: Countdown to 2000 included presentations by TIGR President J. Craig Venter on the state of genomics research, University of Pennsylvania bioethicist Arthur Caplan on genomics' ethical implications, SmithKline Beecham Chief Science Officer George Poste on the healthcare industry's future, and Harvard economist Juan Enriquez on how genomics will alter the global economy. Most addressed the need for various scientific disciplines to collaborate to develop methods for deriving useful information from current genetic databases.
"What genomics is right now is not what we need it to be," Roger Brent of the Institute for Molecular Sciences in Berkeley, Calif., told the audience of some 200 scientists and journalists. "Information is not getting us where we need to go. We'll know the effect of every protein-protein interaction…but that sort of biology by itself will not be predictive," Brent said.
He predicted a transformation in the way biologists practice, and said genomic researchers need to "think harder about data visualization…and data structures." Mathematicians, physicists and other scientific disciplines must be pulled into the field "to help us make sense of the flood of information," Brent said.
Genomics has already transformed the pharmaceutical industry, said SmithKline's Poste. "We're moving from a data-poor environment to one that threatens to swamp us with a tidal wave of data," Poste said. "Where once there was a shortage of targets protected by pharmas, now the challenge is how to choose the target with the highest probability."
Sequence information must be turned into "functional genomics," said Poste. And to do that, "research directors must look well beyond biology to robotics, engineering, computing, optics, and material science," he said.
Brandeis University researcher Dagmar Ringe described the problem as one of getting "from sequence to consequence."
"We need a way to interrogate structures in a way that it will tell us their function," said Ringe.
"The challenge will be met in biological, chemical, biochemical, structural, and computational ways," she said. "We need to work together to achieve these goals."
Poste and others also forecast new relationships between healthcare, computers, and telecommunications. Informatics will drive diagnostics and therapeutics, Poste said. "The framework of genomics will not be able to proceed without an intimate link to informatics."
"We're on the cusp of potential paralysis because not enough attention has been paid to annotation, standards, scale, hyperlinking or costs," Poste warned. "Without a proactive design for system continuity, the consequences will be upon us in two to three years."
Poste criticized what he called "a complete lack of strategic thinking" in genomics. But, as one observer put it, "it's hard to think strategically in a field that is not only brand new, but already is changing quickly."