IBM Announces Funding for Gene Research - Company Business and Marketing

Edge: Work-Group Computing Report, Dec 20, 1999

IBM Research Wednesday announced a grant initiative to support strategic research aimed at understanding the structure of proteins. The million-dollar grants for IBM supercomputing technology will be awarded to select research centers working in the field of structural genomics.

The new grant program complements the $100-million research initiative IBM announced last week to build the world's fastest supercomputer and tackle the grand challenge protein folding problem. Structural genomics research seeks to develop classifications or "families" of protein shapes to enable scientists to deduce the shape and function of a protein based on its gene sequence. Proteins are fundamental to all life, and understanding protein shapes and functions is a key to developing better healthcare.

The IBM grants will be given in coordination with the Protein Structure Initiative of the National Institute of General Medical Sciences, part of the National Institutes of Health (NIH). In June of this year, the NIH announced it would fund three to six pilot centers for research in structural genomics as part of its Protein Structure Initiative. The IBM awards will be given to one to two pilot research centers participating in the NIH program, and will provide funding up to $1 million to each selected pilot center for IBM equipment such as RS/6000 SP supercomputers and other technologies.

"Deep computing technology is dramatically accelerating research in biology, allowing us to shift much of the critical work from the laboratory to the computer," said Sharon Nunes, senior manager of the Computational Biology Center at IBM Research. "This funding program continues IBM's strong tradition of collaboration and support in strategic areas of research like computational biology."

Large volumes of gene sequence data are being collected through the Human Genome Project and other initiatives. Each gene sequence contains the recipe for a protein that serves some function in living systems. The next crucial step for gene research is determining the structure and function of a protein based on its associated gene sequence.

Using computer analysis and modeling, scientists hope to be able to assign each protein to a family based on its gene sequence, and then -- based on understanding of the structure of other proteins in the same family -- generate a reasonably accurate model of the protein structure. Reaching this goal will require determination of as many as 10,000 protein structures -- a task that is only now becoming feasible through developments in technology.

The grants will be coordinated by IBM Research's Computational Biology Center as part of IBM's Shared University Research program, and the pilot research centers will have opportunities to collaborate with the Center and access its advanced software and hardware resources. IBM will host a workshop to outline the resources and opportunities being made available through the award program; further details and guidelines for application will be announced in February, 2000.

The Computational Biology Center at IBM Research IBM's Computational Biology Center was founded in 1997 to bring together scientists and resources in biology, chemistry, physics, and computer sciences for research in computational biology, bioinformatics, and related fields. The Center currently encompasses approximately 30 scientists in five of IBM's eight research labs worldwide.

The Computational Biology Center is part of the Deep Computing Institute at IBM. Deep computing brings together powerful computers and specialized software and mathematical algorithms to attack extremely difficult problems involving large amounts of data. More information on computational biology and deep computing at IBM is available at http://www.research.ibm.com/dci.>

COPYRIGHT 1999 EDGE Publishing

COPYRIGHT 2000 Gale Group