Virtual Human Needs More Than Science

June 10, 2002 (Computerworld)

For researchers, the need for a virtual human may be clearly established, but there are monetary, computational and procedural obstacles.
Although the project has sparked interest at the National Institutes of Health in Bethesda, Md., and the Defense Advanced Research Projects Agency, "the federal agencies don't seem inclined to fund this at this time," says Clay Easterly, a researcher at Oak Ridge National Laboratory in Oak Ridge, Tenn.
He and Charles DeLisi, a professor and director of bioinformatics at Boston University, say they're looking to private companies -- pharmaceutical firms and technology vendors such as IBM -- to fund this research. Pharmaceutical companies already benefit from limited computer models of the heart and liver that are used for drug testing. Those models involve blood flow and metabolic analysis to see how long a chemical or drug stays in the organ.
But pharmaceutical companies don't want to own the project outright. "It's too long-range for them," DeLisi says. The problem with companies is "they have really, really poor judgment. They're too conservative," he adds.
Ultimately, the virtual human needs a sort of foster parent "with money and foresight," DeLisi says. It also needs some governance on how to develop the different models in concert.
Sharing research that's in progress with others -- which is how Easterly has proposed that the virtual human be developed -- conflicts with the way academia works, which is to reward specialists and not collaborative teams, Easterly says.
Researchers are "very focused on reductionist science," Easterly says. That is, they specialize and know a great deal about a very narrow field of study, like how the liver processes chemicals. But they don't know how the processes in the liver affect the rest of the body.
Academia's way of working has been very successful over the years, but now it's almost stymied by the immense amount of information it has produced, say DeLisi and Easterly. To continue, programming and medical research need to progress in concert, DeLisi and Easterly agree. "Our whole process of science has to undergo a revolution before that can happen," Easterly says.