The idea that scientists can be immersed in potential new drug compounds they are studying may sound futuristic, but that's exactly what drug designers working in the brave new world of bioinformatics will be doing through virtual reality.

"You have to be able to interact with your data to learn and discover," said Robert M. Straubinger, associate professor of pharmaceutical sciences in the School of Pharmacy and Pharmaceutical Sciences.

	Robert Straubinger retooled his lab and reworked grant proposals in light of the bioinformatics revolution
	Photo: Stephanie Hamberger

Straubinger, who has been working on a less toxic way to get the cancer drug taxol into tumor cells, actually has retooled his entire lab and reconfigured grant proposals in light of the bioinformatics revolution.

He was intrigued by work conducted by his colleagues, Richard Almon, professor of biological sciences, and William Jusko, professor and chair of the Department of Pharmaceutical Sciences, showing that changes in gene expression could be markers for pharmaceutical effects.

That work involves pharmacodynamics, the study of how drugs affect cells and tissues as a function of time and concentration, which Almon, Jusko and others pioneered at UB.

"In my opinion, pharmacodynamic analysis of gene expression is one of the most profound advances that's come out of bioinformatics so far," said Straubinger.

He is hopeful that he will be able to find changes in genes that relate to the concentration of drug in tumors, potentially a breakthrough because of the difficulty of identifying suitable markers present in both tumors and in normal tissues.

"For most drugs, the effects on cellular gene expression are not known, so gene chips would allow us to scan large numbers of genes for activation," he said.

But Straubinger was confronted with a dilemma when he realized that the amount of data that would be generated by this new approach was far beyond the capability of his lab, or any biologically oriented lab for that matter.

"The problem is that to do this, we need to produce a time course of gene-expression changes in response to several different concentrations of drug in order to see whether different genes turn on and off under different conditions," he said. "So for us to get a good picture of what the drug is doing, we'll be looking at hundreds of thousands of datapoints," he said. "How does a human being comprehend such a mass of data?"

The answer, it turns out, lay a few campus buildings away in the lab of Aidong Zhang, associate professor of computer science and engineering, who studies pattern recognition and multimedia visualization of databases.

She was working on ways of visually representing huge amounts of data in ways that make sense to scientists.

For two years, researchers from the labs of Straubinger, Zhang and Raj Acharya, chair and professor of the Department of Computer Science and Engineering, have been meeting and conducting research together.

The combined research team already has submitted two grant proposals.