SCIENCE

New supercomputer, most powerful at Johns Hopkins, will speed disease research: A federal grant will allow Johns Hopkins researchers to purchase a powerful $2 million computer that will speed up their efforts to find new ways to diagnose and treat brain disease, heart illnesses, cancer and other medical ailments. The Institute for Computational Medicine, based at Johns Hopkins' Homewood campus in Baltimore, will receive one of the 20 High-End Instrumentation Grants for 2008 awarded by the National Center for Research Resources, a part of the National Institutes of Health. The one-time grants, announced today, are awarded to support the purchase of sophisticated equipment costing more than $750,000, machines with the potential to impact a wide range of biomedical research. The $2 million grant to the Johns Hopkins institute was the maximum amount allowed for any single project. Launched in 2005 as one of the first, largest and most ambitious research centers of its kind, the Institute for Computational Medicine focuses on unraveling health problems through methods other than traditional "wet-lab" techniques such as growing cells in a dish. Some of its researchers, for instance, create elaborate computer models that mimic in virtual reality the real-world activity of living cells and organs. Researchers conduct experiments with these models, testing, for example, the effects of experimental medications. Others researchers use information technology to compare digital images of healthy and diseased tissue, looking for early indications of illness. "With this federal grant, we will be able to buy a computer equipped with the next generation of microprocessors, hardware that will be available later this year," said Raimond L. Winslow, director of the institute. "It will be the most powerful computer at Johns Hopkins, and very few other places have a computer this powerful dedicated to solving these types of biomedical problems. The computer will allow us to move toward important discoveries in medical diagnoses and treatment at a much faster speed." The new device is expected to be a 256 dual quad-core node cluster computer with 1 petabyte of storage. A petabyte is a measure of digital information equivalent to 1 quadrillion bytes or 1,000 terabytes. (A common household computer or portable flash drive usually possesses storage capacity measured in gigabytes; a terabyte equals 1,024 gigabytes of data.) The computer will be installed, tentatively in early 2009, in the university's Computational Science and Engineering Building. "We have a 1,000-square-foot room reserved for it, and the computer will fill it up," Winslow said. A vendor for the computer has not yet been selected. The technological resource will be shared by more than a dozen faculty members affiliated with the Institute for Computational Medicine. Winslow predicts that it will also serve perhaps 100 other collaborators from the university's School of Medicine and Whiting School of Engineering and from other institutions. Three Department of Biomedical Engineering faculty members who will be among the institute members using the computer to enhance their research are:

• Natalia Trayanova, who studies how dangerous arrhythmias are initiated and maintained in the heart. The new computer is expected to speed up her efforts to find the best ways to halt these irregular heart rhythms with shocks from a defibrillator.
• Michael I. Miller, who compares the shape of brain structures in images from healthy and diseased patients, looking for differences that may lead to better diagnoses and treatments. Miller now uses linked computers across the country to collect the resources to conduct this research. When Miller gets access to the new Johns Hopkins computer, Winslow said, work that now takes months to accomplish by cross-country connections should take only days to complete.
• Rachel Karchin, who is using computer models to predict how mutations in proteins can trigger the development of breast cancer. The new device should enable her to study this process in more complex and more detailed models, Winslow said.

The Johns Hopkins computer award was part of the $33.3 million in grants allocated for 2008 by the National Center for Research Resources to fund purchases of the latest generation of advanced research equipment. "Innovative biomedical research requires frequent access to the newest and most advanced technology," said center director Barbara Alving. "High-performance equipment provides NIH-funded researchers with new discovery tools enabling a new generation of data and a new dimension of information. Tools such as these play key roles in the study of disease and the fundamental mechanisms of biological function, ultimately leading to new advances and treatments for diseases."