Researcher Finds Biochemist's Holy Grail

Researchers from the Howard Hughes Medical Institute stuck models of the amino acids they found in a protein out into cyberspace, and let 150,000 home computers crunch away at that little tidbit of information. Out from the hat popped a prediction of the structure of that protein, proving that distributed processing and computer modeling can be successfully applied to the complexities of analyzing proteins with nothing but the information about their building blocks.

Why is this a big deal? Until the past six months "...it has seemed nearly impossible to compute the structures of proteins from their amino acid sequences, and solving this problem has been something of a scientific Holy Grail," says Dr. David Baker on the Rosetta@Home site. Baker has grabbed up the cup with the Rosetta@Home application.

Unlike other types of molecules, some proteins are too sensitive to be tested using standard means such as x-ray crystallography, which can destroy the protein and leaving it a mishmash of its components. Additionally, proteins and how they function within a biological system can depend just as much on their 3-dimensional shape as on their component parts, making them highly complex structures with minute energy bonds that can fold and twist as the protein does its job.

Baker, based at the University of Washington, published the results of his Rosetta@Home experiment on October 18, 2007 in the journal Nature, proving that the concept first advanced by the Folding@Home experiment can bear fruit in the world of complex protein structures.

The basic idea behind Rosetta@Home and Folding@Home is to harness the downtime of the hundreds of thousands of personal computers around the globe, by applying their computational capabilities to complex modeling programs. Each user downloads the basic algorithm, and their computer receives on a regular basis a new set of problems to work on, and sends the models that individual machine created back to the main computer. The program operates in the background, using snippets of the CPU's time whenever the computer isn't in use.