Thousands of personal computers volunteered by their owners helped scientists solve the mystery of how a protein gets its shape, according to a study published in the British journal Nature.

The computing power was used by researchers led by Vijay S. Pande at Stanford University to find out, atom by atom, how a protein molecule folds into its knotlike form, the study said. The information will be used in the study of Alzheimer's disease, the human form of "mad cow" disease and other illnesses related to protein structure.

"There's a huge social aspect to this," Charles Brooks, who uses supercomputers to carry out similar simulations at the Scripps Institute in San Diego told Nature. "It's not a traditional way to do science."

Solving the puzzle took the equivalent of 2,000 years of computing time. Pande's project, called Foldinghome, joined 200,000 volunteers, each of whom downloaded computer software that went to work on the protein puzzle when the computer was idle, Nature reported. The software included a swirling screensaver and a data-analysis program. The results from each computer were sent to the university over the Internet.

The concept, called distributed computing, has been used since 1999 in the search for extraterrestrial intelligence by the SETIhome project, based at the University of California at Berkeley and involving 4 million personal computers.

Pande, who recognized that research into proteins "is not very sexy" compared with the search for aliens, spurred interest in his project by setting up a competition to see which PC user could donate the most computer time, Nature said.

PC users for the protein study were recruited through newspapers and Web sites.