Taking the Bioterrorism fight to home PCs

Whenever you're not doing anything else on the machine -- even during the split-second between two keystrokes -- the software will run its chemical analysis. All you have to do is log onto the Internet once in a while,

Here's one way to strike a blow against bioterrorism: leave your PC running. Texas computer entrepreneur Edward Hubbard says it'll help. So does Graham Richards, chairman of the chemistry department at Oxford University. They're two of the people taking part in a digital assault on smallpox, which could be used by terrorists to kill millions.

What's needed are better drugs to treat diseases that could be unleashed through biological attacks. Finding these drugs will require a lot more than lab rats and test tubes. The scientists also need computing power, quite a few supercomputers' worth, in fact. And that's where you come in.

Here's the plan: Just leave your computer running. But first go over to United Devices, at www.ud.com, and download a little piece of software. It's a program that hunts for chemicals that might disrupt the reproductive cycle of the smallpox virus.

Install the software, and leave it running. Whenever you're not doing anything else on the machine -- even during the split-second between two keystrokes -- the software will run its chemical analysis. All you have to do is log onto the Internet once in a while, so the program can send its results back to headquarters, and download a new batch of data to analyze.

Your desktop computer is now part of a worldwide network called the PatriotGrid, which has already attracted a million members running 2 million PCs. Hubbard, chief executive of United Devices, says the smallpox project would take about 45 years on a scientific computing cluster with a thousand processor chips inside. With PatriotGrid, he expects to get it done in a couple of months.

''We basically have a way for people who are concerned about bioterrorism, not only in the United States but all over the world, to plug into a network and actually do something about it,'' said Hubbard.

United Devices isn't doing this solely out of patriotism. It hopes to demonstrate the value of ''grid computing'' systems, which let companies use their surplus processing power to solve heavy-duty computing problems, like calculating actuarial tables.

The most famous grid computing system is the SETI@Home project at the University of California at Berkeley, which uses thousands of Internet-based computers to look for life on other planets.

But United Devices set a more practical goal, teaming up with Oxford University to find new cancer drugs. The war on terrorism gave the company new targets. It has already completed a project to look for possible anthrax cures. Now the system is working on smallpox. There's a vaccine against this disease, but no cure for those already infected.

To find one, scientists are working at the molecular level. Cells, like everything else, are made up of billions of chemical molecules, some of which can interact with neighboring molecules. These ''active sites'' on a cell are like keyholes in a door. If another molecule of the right shape comes along, it fits into the active site like a key. This combination can cause the cell to react in a particular way. An effective drug is really a batch of molecules that fit the active sites on certain cells, altering the cells' functions.

When scientists figured this out three decades ago, it was as if cavemen armed with stone axes had discovered bronze. The hit-and-miss guesswork once common in drug development was largely abandoned. A scientist hunting for a new cancer drug can immediately rule out billions of candidates that couldn't possibly fit the active sites on a tumor, and focus on the billions that might work. And because molecules can be simulated in a computer, the scientist can test likely drugs in silicon rather than than a petri dish.

Still, we're dealing here with billions of possibilities. Even if the biology lab has a Cray supercomputer in the basement, it would take years to do all of the number crunching needed to develop a single medicine. And because the Cray is relatively slow, it would test a far smaller set of potential cures. Grid computing produces superior results, said Oxford's Richards, because it's fast enough to test a lot more molecules in a shorter period of time.

''This is just doing what is already done,'' he said, ''but better and faster.''

At the end of the project, United Devices and Oxford University will have a list of thousands of molecules that bond to the smallpox virus and might help kill it. The data will be handed over to the Department of Defense, which is trying to develop a smallpox cure. Even with the help of the PatriotGrid, a cure for smallpox is years away, but Richards estimates the grid will shorten the process by as much as two years.

As an antiterrorism strategy, PatriotGrid lacks the visceral satisfaction of dropping a bomb on Osama's cave du jour. But it might help save a few lives someday -- or maybe a few million.

Hiawatha Bray can be reached at bray@globe.com.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.