2005 Computerworld Honors Program

	Supercomputer simulation of enzyme DNA interaction Pittsburgh Supercomputing Center Pittsburgh, PA USA Year: 1993 Status: Award Recipient Category: Science Nominating Company: Cray Research, Inc.
	Interaction between proteins and DNA is a fundamental biological process. Using supercomputing to simulate this process is an endeavor in basic research that benefits society by creating greater ability to control human disease and improving the quality of life.
	The discovery of the double helix structure of DNA, worked out in its rudiments by Francis Crick and James Watson in 1952, is without question one of the scientific triumphs of this century. DNA is the basic repository of genetic information, an amazingly intricate molecular code governing the biological processes we call "life." Nearly everything that happens biologically from DNA happens because many different proteins somehow have the ability to isolate their activity at specific sites on the long helical strands that comprise DNA molecules. John Rosenberg uses the CRAY Y-MP supercomputing system at the Pittsburgh Supercomputing Center to understand these complex processes, usually called protein-DNA recognition. "There are large classes of proteins," says Rosenberg, "that recognize specific sequences of DNA, and they do very important things - ranging from controlling which genes express when and how, to rearranging the structure of DNA itself. These are basic biological processes that we want to understand partly for their own sake, but also for two practical reasons: first, many disease processes may be related to aberrations in these events and, second, restriction enzymes are vital tools of the biotechnology industry." Nature's Scalpels: Restriction Enzymes For 15 years, Rosenberg and the scientists he works with at his University of Pittsburgh laboratory have studied the DNA-recognition mechanisms of a protein called Eco Rl endonuclease. Eco Rl is one among a class of enzymes, called restriction enzymes, that protect the DNA in bacterial cells by attacking DNA brought in by viruses and other foreign agents. "Eco Rl is one of the most frequently used in what's called recombinant DNA technology," says Rosenberg, "or in the Jargon 'cloning.' These enzymes recognize a particular sequence of bases of DNA and cut the DNA at those sites, breaking it into well-defined pieces that can be put back together in new combinations. Eco Rl is the prototype, the first one of these enzymes to be understood and the first one used in this technology. The CRAY Biology Lab Rosenberg and his graduate students, now Ph.Ds, John Grable and Yongchang Kim used X-PLOR, an efficient program for refining the structure of biological molecules (developed by Axel Brunger of Yale), to help identify the mechanisms by which Eco Rl recognizes a particular sequence of DNA. These computations resulted in clarification's to the structural model of Eco Rl and how it binds to DNA. Their computations showed that the protein wraps almost completely around the DNA, as if embracing it with extended arms, and at the same time kinks the DNA at the site where it binds. Part of the structure resembles the structure of other enzymes that bind to "nucleotides." Rosenberg believes that this portion of the structure, called the nucleotide-binding fold, may be one of the keys to understanding protein-DNA recognition. "This architecture is connected very deeply to how proteins recognize nucleic acids. In an evolutionary sense, it's one of the ancient patterns." Large-Scale Molecular Dynamics With recent physics Ph.D. Shankar Kumar and Peter Kollman of the University of California at San Francisco, creator of the AMBER molecular dynamics program, Rosenberg designed Y-MP simulations of the kinked DNA to examine whether the kink was intrinsic to the DNA or was caused by binding with Eco Rl. "The computations gave us a very definite answer," says Rosenberg, "that the kink results from binding with the protein, and this leads to a whole series of ramifications about how the system works." In other simulations with AMBER, Rosenberg and Kumar stretched the limits of computational biology. Their simulations of DNA in solution (with 1970 water molecules for 81 picoseconds) represent one of the most extensive molecular dynamics computations on DNA molecules to date.