Stanford University
Protein Researchers Bring Mac Users into the Fold
Teamwork on a Massive Scale
Previous efforts to simulate protein folding have been unsuccessful because there was no way to efficiently divide up the work among tens of thousands of computer processors. “Having 10,000 assistants would not be helpful, for example, because it takes too long to divide the tasks and tell each assistant what to do,” Dr. Pande explains. But Dr. Pande has developed a new way to simulate protein folding called distributed dynamics that removes previous barriers to dividing up the work among large numbers of processors.
The Pande Group already has about 20,000 PC users who have downloaded the software. Volunteers do not need to have a fast Internet connection, because according to Dr. Pande there is only a small amount of data that travels back and forth between remote client machines and the PowerBook G4 at Stanford. Periodically, the client software connects up to transfer data, as soon as a specified amount of processing is completed. The client software allows volunteers to specify how much their machine is used, as little or as long as they like. “Even a single day’s worth of computing is helpful to us,” says Dr. Pande.
“Creating Mac OS X versions of our existing UNIX-based client and server software was painless and fast — a matter of a few days.”
— Dr. Vijay Pande
A Contribution to Science
Macintosh users who want to donate idle computer time need only download the software from http://folding.stanford.edu/download.html and install it. As the software runs in the background, a window provides users with a graphical view of the protein that their computer is helping to fold, as it is folding. According to Adam Beberg, who brought the Folding@home client and server software to Mac OS X, Macintosh users should enjoy using the software. “We used OpenGL to create the graphical interface for Mac users, which took only about two days,” says Beberg. “Mac users really appreciate the slick, clean interface.”
While it’s too early to talk about most of the proteins that have been folded so far, Dr. Pande says all results will be open and published. Some of the early work of the project has already been published in leading scientific publications, including Science, Physical Review Letters, and The Journal of Molecular Biology. One thing is certain: to learn what they need to learn as quickly as possible, Dr. Pande and his associates are calling the help of Mac users everywhere. “We’re excited about getting a strong showing from the Mac community,” says Dr. Pande.
