Home computers add enormous power to drug discovery programs.

OK computer: The screensaver project cuts years from the drug discovery process.

Sure, your computer can send e-mails, create spreadsheets and assemble PowerPoint presentations. But can it save lives? Any computer with access to the Internet can now help discover cancer drugs by running a special screensaver.

Launched in 2001, the Screensaver-Lifesaver program uses parallel computing power and virtual screening to assess the interactions between small drug-like molecules and predetermined cancer-causing targets.

More than 3 million personal computers worldwide are screening a library of 3.5 billion molecules against these targets to identify potential candidate drugs. This may shave up to 3 years off the drug discovery process, researchers say.

The screensaver project can us save huge amounts of time. , Daniel Von Hoff, Center for Targeted Cancer Therapies

“What the screensaver adds to the discovery process is an enormous amount of computer power, dwarfing what even the biggest pharmaceutical company can do,” says Graham Richards, chairman of chemistry at the University of Oxford. “In our hands it provides sheer power to do things that would be beyond the capabilities of even the biggest and most costly machines.”

The latest scheme, launched in April, is testing several new protein targets—primarily kinases and phosphatases—for pancreatic cancer.

Each computer receives the screensaver program, which includes drug-design software, an initial packet of 100 molecules and a model of a target protein. The program calculates the binding energy between the small molecules and the targets. Molecules with the tightest binding have the best chance of becoming drug formulations because tighter bonding translates into fewer adverse effects and lower doses. Once processing is complete—typically in a day—the program sends the results back to a data center and requests more molecules.

“Using our best models and guesses, it would take hundreds of years of trial and error in the lab to test these protein targets,” says lead investigator Daniel Von Hoff, director of the National Foundation for Cancer Research's Center for Targeted Cancer Therapies in Arizona. “The screensaver project can us save huge amounts of time and find the most optimal chemical structures—those that bind the tightest.”

In addition to searching for cancer therapies, researchers are using the program to find drug targets for smallpox and anthrax. In less than four weeks, the Anthrax Research Project found 376,064 potential candidates for new antianthrax drugs; 12,000 of those are being investigated.

Another screensaver project aims to predict the structure of proteins found in the human genome. The structures of only an estimated 30% of proteins encoded by the human genome are known. The Human Proteome Folding Project is deciphering the three-dimensional structure of human proteins with no known structural homologs, proteins made by pathogens and those encoded by genomes of environmental microbes. The project is slated for completion by the end of 2005.

“This is an example of a project so big that no one thought it was solvable,” says Ed Hubbard, President of United Devices, the company that powers the program's computing platform. “But it is possible and it's changing the way researchers think about problems.”