January 31, 2002-Vol33n15: NSF grant to fund data storage system

VOLUME 33, NUMBER 15

THURSDAY, January 31, 2002

NSF grant to fund data storage system
$1 million award will boost pioneering research in bioinformatics and GIS

By ELLEN GOLDBAUM
Contributing Editor

A $1 million National Science Foundation infrastructure award to store, manage and analyze complex scientific data is boosting pioneering research at UB in bioinformatics, geographic information science and other important research areas.

Nearly $600,000 in matching funds will be contributed to the project by UB and the Strategically Targeted Academic Research (STAR) Center for Disease Modeling and Therapy Discovery, funded by the New York State Office of Science, Technology and Academic Research.

The highly competitive grant provides UB with the computational infrastructure necessary to manage, analyze and visualize large-scale, multidimensional data sets that lie at the heart of some of the university's most cutting-edge research in drug design, molecular-structure determination and the understanding of complex geographic images.

"The timing of our receiving this grant now from the National Science Foundation is key," said Jaylan S. Turkkan, vice president for research.

"In conjunction with our new Center of Excellence in Bioinformatics, it will ensure that UB is at the leading edge, not only in high-capacity storage of large datasets that form the knowledge base for bioinformatics, geographic information science and other important research areas, but also in developing the complex software tools that will be needed for understanding and analyzing the data.

"The combination of these two events is a one-two punch in terms of leveraging UB for large-scale center grants from the National Institutes of Health and other federal agencies," Turkkan added.

The new system will provide the Department of Computer Science and Engineering with 20 times the amount of data storage it now has in a centrally managed resource that will assist affiliated researchers working in the departments of Geography and Pharmaceutical Sciences, and at Hauptman-Woodward Medical Research Institute and Roswell Park Cancer Institute (RPCI).

In emerging disciplines, such as bioinformatics and geographic information science (GIS), the issue of data storage has shifted dramatically from one easily solved by filing cabinets and the hard drives of PCs to one that can be solved only by extremely specialized computer equipment.

"The bottom line is, if you cannot store the data, you cannot do the research," stated Aidong Zhang, associate professor of computer science and engineering, and principal investigator on the grant.

Research projects that are named on the grant include:

Determination of the 3-dimensional structure of proteins (Russ Miller, professor of computer science and engineering, and director of UB's Center for Computational Research)

Metadata and knowledge extraction, representation and management in geographic information science (David Mark, professor of geography and director of the UB branch of the National Center for Geographic Information and Analysis, and Zhang)

Gene-expression data analysis (Robert Straubinger, associate professor of pharmaceutical sciences; Murali Ramanathan, associate professor of pharmaceutical sciences, and Norma Nowak, director of the DNA Microarray Facility operated by RPCI and the School of Medicine and Biomedical Sciences)

Data visualization (Ashim Garg, assistant professor of computer science and engineering, and Raj Acharya, now at Penn State)

Data management (Zhang)

By the end of the five-year grant, UB will have a total of 20 terabytes of computer storage space. A single terabyte is 1 million megabytes. Twenty terabytes is approximately the same amount of storage found in 1,000 PCs, although researchers caution that the new system is far more than just a fantastically large storage space.

"It's not just this huge amount of storage. It's high-performance storage that enables large amounts of data to be moved around and accessed quickly and easily," said Straubinger.

Zhang pointed out that in addition, the new system will be extremely reliable—another prerequisite for storing such vast amounts of data, since losing so much data would be catastrophic for any research project.

"Data is written in multiple places in the system, so that a hardware failure won't result in data loss," she said.

The new infrastructure also will facilitate the kind of computational research required by these emerging disciplines.

"In pharmacogenomics, for example, there is a tremendous need for computational research focused on analysis of the kinds of data we generate, with the objective of better understanding the mechanism of drug action or how cells and tissues respond to drugs," said Straubinger.

Zhang, whose expertise is in database management and data mining, is working with other UB researchers to develop just those tools.

Having these machines on campus, she explained, will enable UB computer scientists to begin to develop the data analysis and visualization systems that will make working with such large data sets more efficient.