Thursday, October 23, 2003
University of Buffalo Reporter, Volume 35, Number 9
Like a puzzle with key pieces missing, gene-expression data provide medical researchers with a critical, but incomplete, story of how complex diseases develop. The other parts of the story only can be revealed when genomic data are analyzed together with clinical and epidemiological data.
Linking computer scientists with life scientists to develop computational tools that will help draw a far more complete picture of the causes behind complex diseases like cancer, multiple sclerosis and coronary artery disease is the goal of researchers at UB working under two major federal grants totaling $2.8 million.
It's the kind of comprehensive approach that many say is critical if significant research progress on these diseases is going to be made.
A $1.2 million grant from the National Institutes of Health will establish a Planning Center for Biomedical Computing, where biomedical scientists and clinicians will work alongside computational scientists, developing "real-world" techniques for storing, managing, analyzing, modeling and visualizing multi-dimensional data sets that describe complex diseases.
The second grant, $1.6 million from the National Science Foundation, will fund use of computational methods to integrate relevant genomic data into the many different kinds of clinical data that exist on thousands of patients diagnosed with cancer, heart disease, MS and other chronic conditions.
"This is exactly the wave of the future, creating close collaborations between experts in genetics, public health and computer science who can look at an issue together and try to solve it," said Maurizio Trevisan, acting dean of the School of Public Health and Health Professions and a co-principal investigator on both grants.
Trevisan noted that UB's significant strengths in computer science, particularly in applications to life-sciences problems, position the university as a leader in this "next wave" of medical research.
Kenneth Tramposch, associate vice president for research, said "there is a pressing need to develop computational methods that can analyze the very large databases that the life sciences are generating, so these close collaborations that we are developing, where computer scientists will actually be assigned to life sciences labs, are critical."
The result will be to advance the understanding of disease and optimize drug therapy based on pharmacogenomics—how certain genes respond to drugs and how those genomic responses are, in turn, influenced in patients by immune, hormonal, environmental and other stimuli.
"These grants are about clinical bioinformatics," said Aidong Zhang, professor of computer science and engineering in the School of Engineering and Applied Sciences and principal investigator on the two grants. "Our goal is to develop computational and visualization tools to integrate data from population studies and clinical data, such as results of lab tests, MRI tests and others with pattern-analysis results on the genomic data, allowing medical scientists to more easily discover the meaningful connections between the two."
According to Trevisan, it is those connections that hold the keys to the causes, and ultimately the cures, for these chronic diseases.
"Most recently, genetic studies in cardiovascular disease have focused primarily on single genetic variations, but to date these have not been very successful," he explained. "That's because it's not just a single genetic variation that causes the disease, but rather multiple genetic causes that interact with the environment and with other individual characteristics. That's where it becomes really complex and nobody has found yet a very efficient way to do this. But if we want to explore the contribution of genes and individual characteristics to the burden of disease in the population, this is what we need to do."
UB brings to these emerging problems a special breadth of expertise because of its schools of medicine, engineering, pharmacy and public health, noted Jaylan S. Turkkan, vice president for research.
"On a national level, the application of computational tools to solve biological and medical problems has been a top priority with the federal agencies," she said. "With momentum building in the UB Center of Excellence in Bioinformatics, our new School of Informatics, and this new Planning Center for Biomedical Computing, UB is now a force to be reckoned with for future funding."
Murali Ramanathan, associate professor of pharmaceutical sciences in the School of Pharmacy and Pharmaceutical Sciences and a co-investigator on both grants, said the projects address what researchers refer to as the "over-availability of data and the under-availability of knowledge," where the enormous volume of data is of little value without techniques to "mine," or make sense out of it.
"The tools we are developing will allow researchers to do genomic data analysis in a clinical context," he said.
With a disease like multiple sclerosis, Ramanathan explained, sophisticated data-mining tools will prove invaluable if they can identify how patient characteristics—gender, age, the type and level of disability, and the form of disease—influence genomic variables and how they, in turn, contribute to the patient's experience with the disease.
The grants build on key strengths at UB that include large epidemiologic databases on chronic diseases gathered by researchers in the School of Public Health and Health Professions, as well as access to the nation's largest registry of MS-patient data, the New York State Multiple Sclerosis Consortium Database, gathered under the direction of the late Lawrence Jacobs, a pioneer in MS research who was chair and professor of neurology at UB.
Both grants involve multidisciplinary groups of investigators with expertise
in data mining, cluster and pattern analysis, gene-environment interactions,
mathematical models for pharmacogenomics, molecular mechanisms of disease,
neurology and neuroimaging from the School of Engineering and Applied Sciences,
School of Medicine and Biomedical Sciences, School of Pharmacy and Pharmaceutical
Sciences, and School of Public Health and Health Professions, as well as
Roswell Park Cancer Institute.
Copyright © 2003, University at Buffalo