Subscribe to Online or ENVISION | Contents | Cover

Supercomputer Shows Role of Dynamics When Enzymes Link to Substrates

It has long been known that enzymes recognize substrates depending to a very large extent on the shape, or conformation, of both molecules. The substrate fits like a hand wriggling into a pocket in the enzyme, called the active site. A far-flung team of scientists has shown that shape isn't everything: the wriggling counts, too. With large-scale computer simulations run at SDSC, they've shown, quantitatively, how one of the fastest enzymes in the world does its work.

Huan-Xiang Zhou (Drexel University and Hong Kong University of Science and Technology), Stanislaw T. Wlodek (University of Houston), and J. Andrew McCammon (UC San Diego) combined computational models and theoretical calculations to obtain their results, which are published in the August 4 Proceedings of the National Academy of Sciences.

The team studied the enzyme acetylcholinesterase (AChE), which controls the communication among nerve and muscle cells. The speed of AChE has been puzzling, because its active site appears to be accessible from the enzyme's surface only by a partly blocked channel. Earlier work by the same team showed that "breathing" motions in AChE open and close the channel to allow the substrate acetylcholine (ACh) to enter the active site.

The new work shows that the breathing motions allow ACh to bind almost as fast as if the channel were always open. The work also shows that the motions prevent the binding of possible substrates that are larger than ACh. The result is a new concept, dynamic selectivity, for recognizing molecules in non-equilibrium systems such as living organisms.

NPACI Computer Compares More than 8,000 Protein Structures; Database Already Turning Up Discoveries

A biologist might never think to compare a snake toxin and a harmless bacterial enzyme involved in nitrogen transfer. However, a database created by biologists at SDSC is already turning up discoveries such as a structural feature shared by these proteins. The uncovered feature may provide clues to some distant evolutionary relationship between the proteins, or at least an energetically favorable arrangement repeated in nature.

In the September 1998 issue of Protein Engineering, SDSC researchers Ilya Shindyalov and Phil Bourne report on the Web-accessible database they created that compares the structure of more than 8,000 publicly available protein structures to each other. They built the database, which would have taken a year and a half on a smaller computer, in several weeks on SDSC's CRAY T3E.

"The result is a database of structure neighbors--proteins with varying degrees of structural similarity," Bourne said. "This tool is a valuable community resource in our quest to understand more about biological systems." The database is supported by SDSC's Biological Data Representation and Query initiative and NPACI's Molecular Science thrust area.

While there are many methods for comparing protein structures, the SDSC database is one of the few Web-accessible resources with calculated comparisons. The site includes alignment and Java-based visualization tools to investigate a protein's structure neighbors. Users may also submit structures not already in the database and get comparison results via e-mail.

CAIDA Receives DARPA Award to Lead Effort to Measure, Predict, and Engineer Internet Infrastructure

Under the auspices of the Next-Generation Internet initiative, DARPA awarded a two-year cooperative agreement to the Cooperative Association for Internet Data Analysis (CAIDA) for a research effort based at SDSC.

The project represents a leading-edge technology effort to advance the technology and infrastructure to measure, visualize, and predict traffic behavior on the Internet and on advanced networks such as the Next-Generation Internet (NGI). Project researchers will develop and deploy tools to better engineer and operate networks, to enhance network security, and to identify traffic anomalies in real time.

Founded with seed money from the NSF and based at SDSC, CAIDA is a non-profit research organization established in 1997 to promote greater cooperation in the engineering and maintenance of a robust, scalable global Internet infrastructure.

"Understanding macro-level traffic behavior at the infrastructure level is fundamental to our ability to manage and scale the Internet," said K.C. Claffy, principal investigator of the project and founder of CAIDA. The proposed tasks represent areas of high priority to networks, but which are beyond the purview of individual companies.

Japanese and U.S. Scientists Succeed in Trans-Pacific Operation of Ultra High Voltage Electron Microscope

American and Japanese researchers cooperated in a successful demonstration of trans-Pacific telemicroscopy in which the world's largest and most powerful transmission electron microscope (3-million volts)--in Osaka, Japan--was operated from UC San Diego. As part of the demonstration, both American and visiting Japanese scientists at UC San Diego viewed neurological specimens of the sort used in research on Alzheimer's disease and Parkinson's disease.

On June 25, researchers at the National Center for Microscopy and Imaging Research (NCMIR) at UC San Diego used a computer system developed by the Research Center for Ultra High Voltage Electron Microscopy (UHVEM) in Osaka and Hitachi Ltd. to operate the microscope located at Osaka University.

Prior to the demonstration, the NCMIR lab sent two specimens--spinal ganglia neurons from a frog and neurons with spiny dendrites from a rat--to Japan. The NCMIR scientists examined video images of these specimens transmitted in real time by satellite from the Osaka microscope to San Diego while they controlled the microscope by issuing commands with a workstation at NCMIR.

New AlphaServer Continues SDSC's Partnership with Digital Equipment Corporation

SDSC acquired two new AlphaServer 4100s as part of a research agreement with Digital Equipment Corporation to evaluate the platform in a variety of biochemistry, graphics, and data-intensive applications. The two AlphaServers are the first phase of a larger acquisition that will cluster 16 processors in four nodes.

SDSC will evaluate the viability of the AlphaServer 4100 cluster, also known as Rawhide servers, as a high-end server in a supercomputer environment and report the results to Digital's high-performance division. Each of SDSC's AlphaServers will have four 533-MHz Alpha 21164 processors and 2 GB of memory, for a peak performance of 8.5 Gflops. The performance of the new cluster was a key factor in SDSC's decision to replace its older Alpha Farm with the new hardware.

SDSC Vislab Upgrades with SGI Onyx 2 and Octane Workstations

SDSC's Advanced Scientific Visualization Laboratory (Vislab) recently upgraded its graphics capabilities with five new graphics systems from Silicon Graphics, Inc. (SGI), including a four-processor Onyx 2 and four Octane workstations.

The Onyx 2 has four 250-MHz R10000 processors, 2 GB of memory and 81 GB of disk storage. The new Onyx 2 will complement the Vislab's existing four-processor Onyx system. The new machine will handle background processing jobs, remote video serving, and scientific visualization tasks.

For interactive graphics processing, the Vislab has replaced four of its older workstations with SGI Octane systems. Three of the workstations have 250-MHz R10000 processors, MXE graphics boards, 256 MB of memory and 9 GB of disk space. A fourth Octane system has a 250-MHz processor, an MXE graphics board, 512 MB of memory, 45 GB of disk storage, as well as digital video and digital audio processing boards. This system will be used to perform high-resolution video compositing operations over thousands of video frames.

Award to Trace Center Will Help Make Technology Universally Accessible

The Trace Center, of the College of Engineering at NPACI partner the University of Wisconsin, Madison, has been awarded a five-year grant from the U.S. Department of Education's National Institute on Disability & Rehabilitation Research (NIDRR) to make information technology more accessible to people with disabilities. The Trace Center is participating on the Access & Inclusion team of EOT-PACI.

"Universal Design practices based on this research will help make the NPACI infrastructure more effective in reaching more people," says Gregg Vanderheiden, director of the Trace Center.

The NIDRR funding supports students and researchers exploring new approaches for making next generation information technology usable by everyone.

NPACI Earns High Marks on First User Survey

To prepare for the second year of operation, NPACI Consulting conducted its first user survey to assess satisfaction with the partnership's resources and services. The results indicate that NPACI earns high marks from users in most areas.

Users rated NPACI in the areas of computational and archival resources, allocations, applications, consulting, documentation, and training. When it came to availability, users wanted more. "As one of only two NSF Centers serving the entire country, NPACI should have far more powerful machines," one user stated. "It should provide its users with machine capabilities comparable to those available at the DOE labs."

When asked about additional resources, 76% of respondents said more CPU time would be beneficial, 67% said more disk storage, and 47% said more memory. Faster network connections would also increase productivity for many users.

"We want to make sure that we continue to give user needs top priority," said Jay Boisseau, manager of the SDSC HPC Consulting Group. "Given the number of resources the partnership now supports, we're quite pleased with the results of this survey." Surveys will be conducted yearly to address user needs.

Subscribe to Online or ENVISION | Contents | Cover