Faculty Spotlight: Alex Tropsha, PhD
Alexander Tropsha conducted research in Moscow for several years during the 1980s. That experience helped him develop an appreciation for the research setting that awaited him when he came to the United States.
“I think the biggest difference is the potential for growth, the sense that I am not restricted by resources,” says Tropsha, who left his homeland for the U.S. shortly before the Soviet Union collapsed. “I am basically restricted here only by my intellectual abilities, and I had much more restrictions back in Russia for a number of reasons. So I don’t think I would have fully realized my potential being in the Soviet system.”
Tropsha’s intellectual abilities have served him well in the U.S., where he started as a postdoctoral fellow at UNC-Chapel Hill in 1989. Today, he is the associate dean for research at the UNC Eshelman School of Pharmacy, and thanks to an innovative program from the National Institutes of Health, he is pursuing two research projects that combine the latest technology and scientific strategies in an effort to improve medical research in the twenty-first century.
Tropsha has received two NIH Roadmap grants. The overarching goals of the Roadmap are to expand our understanding of biology through interdisciplinary collaboration and to encourage rapid medical improvements that can be translated quickly to patient care.
In 2005 Tropsha received a two-year planning grant from the NIH to create the Carolina Exploratory Center for Cheminformatics Research. The grant, which totals $762,000, involves twelve other labs from UNC and other universities. In 2006 Tropsha received a three-year, $1 million grant for a project focusing on work in his own lab.
Tropsha, who is principal investigator of both projects, wrote the grants in close collaboration with Weifan Zheng, PhD, a 1997 graduate of the School who is now an associate professor at North Carolina Central University.
The Rise of Cheminformatics
Both grants deal with the field of cheminformatics, which Tropsha describes as “a discipline where information and informatics methodologies are applied to storing, managing, exploring, and exploiting chemical databases.” Basically, cheminformatics combines chemistry and computer science to aid in drug discovery.
In the School, Tropsha directs the Laboratory for Molecular Modeling, which is entirely computer based. He says the main pursuit of his lab is understanding the relationship between chemical structure and its function and using that understanding to make testable predictions of which molecules are potential drug targets.
To conduct this research, he must have access to biological data—the results of experimental testing of various molecules. The data is compiled in databases of chemicals—such as a recently established resource at NIH called PubChem—that have been tested against a variety of biological targets. The models developed in the Tropsha group are used to screen untested chemical libraries to identify the molecules that will make the most likely drug candidates.
Tropsha says that in the past, it was not unusual for an experimental lab to spend a year synthesizing and testing a few dozen molecules. Thanks to the development of high-throughput methodologies, labs can now test hundreds of thousands of molecules in a year—much too many for one person to evaluate manually.
He says that’s why the field of cheminformatics was born. While it has been a domain within the pharmaceutical industry for some time, the high cost formerly associated with high-throughput technology limited its application in the academic setting. Now that the technology is less expensive and more readily available, more and more academic labs have started producing large amounts of data, creating a need for the study of cheminformatics in an academic setting.
Ultimately, Tropsha’s lab helps experimental scientists select specific molecules to test in an effort to save time and money.
“The outcomes of my research are models that help experimental scientists make choices,” Tropsha says. “It essentially has evolved to become a decision support system for experimental scientists to plan the next round of experiments and get to the point of discovering potential drug candidates much faster and in a much more economical way.”
Tropsha says he was initially drawn to the field of computer-assisted drug design because of the data-analysis aspect of the work. When he applied his analytical skills to the analytical tools at his disposal, however, he found that the tools had major shortcomings.
“I started by using existing tools, and in doing so, I rather quickly discovered that the tools are not comprehensive and not that accurate,” he says. “And that led us to start discovering tools and methodologies that specifically emphasize accuracy of prediction.”
Both of Tropsha’s grants were part of the Molecular Libraries initiative within “New Pathways to Discovery,” one of the major themes of the Roadmap. Tropsha says that the driving force behind this theme is the recent completion of the human genome sequence and the desire to understand how the functions of every gene can be modulated.
To achieve this, the Molecular Libraries initiative stresses the importance of interdisciplinary collaboration and the use of sophisticated computational techniques and informatics.
Tropsha’s planning grant to establish the Carolina Exploratory Center for Cheminformatics Research involves the collaboration of researchers from twelve labs at UNC and three other universities. These include labs in computational chemistry, chemical biology, data mining, computer science, and statistics.
“The cheminformatics center grant is unique in the sense that it asks to promote an entirely new discipline and organize various research labs with somewhat complementary research interests around a specific new area of research that was only formulated in the scientific literature less than ten years ago,” Tropsha says.
The center will involve developing technical strategies and creating an administrative infrastructure to address critical issues in cheminformatics, which Tropsha says is no longer simply a computational methodology, but rather an entire scientific discipline unto itself. The center’s focus is on establishing a collaborative environment among the participating laboratories, investigating multidisciplinary approaches to key cheminformatics issues, and building a prototype of a Web-based Carolina Cheminformatics Workbench, or C-ChemBench, that can support scientists by allowing them to mine available chemical and biological data to design new compounds or compound libraries with better success.
Within the next year, the Roadmap is expected to issue a request for proposals for funding full centers.
While the first grant involves collaboration with a dozen labs, Tropsha’s second grant focuses on work being conducted in his lab but also involves collaboration.
The project seeks to establish a universally applicable predictive framework for identifying important properties of molecules that determine whether they are viable drug candidates. These properties include absorption, distribution, metabolism, excretion, and toxicity. The ultimate goal of the project is to share modeling software and specialized predictors with the research community through a Web-based portal.
The project involves a research team of investigators with skills in computational drug discovery, experimental toxicology, statistical modeling, and software development and integration.
“I religiously believe in collaborations, and both of these grants are of a collaborative nature,” Tropsha says. “My scientific life is impossible without collaborating with extraordinary bright people who have either similar or complementary research interests to mine — computer scientists on one side, biologists and pharmacologists on the other.”
Tropsha says the Roadmap program has helped him become a better researcher.
“I think it pushed me to do more challenging research—to think about becoming less inwardly theoretical in a way—and establish more meaningful collaborations,” Tropsha says.