Computer algorithms developed for fighting spam are being used to fight a different enemy - HIV, the forerunner to AIDS.

Among a team of scientists given a presentation at an upcoming AIDS conference over software models used to uncover patterns in HIV are two Microsoft computer scientists, David Heckerman and Nebojsa Jojic.

The pair are collaborating with bio-engineers from Australia's Royal Perth Hospital and the University of Washington in Seattle to examine HIV's wild mutation patterns, a better understanding of which researchers see as a key step toward developing broadly effective AIDS vaccines.

Microsoft uses complex data-mining tools to aid Outlook and Hotmail is recognising spam from incoming e-mail. With spammers constantly adjusting their messages to beat automated filters, spam-detection tools dynamically evolve to uncover changing patterns.

The catalyst for the alliance between Microsoft's researchers and medical scientists was the idea that software designed to link "VIAGRA" and "V1AGXA" might also be adept at tracking DNA sequence mutations. If scientists can find stable sequences that persist through multiple HIV strains, they can more effectively craft vaccines to target those areas.

Simon Mallal, executive director of the Royal Perth Hospital's Centre for Clinical Immunology and Biomedical Statistics, credits Microsoft's technology with enabling the medical research team to sift through patient data 10 times faster than any previous research technique.

The group's vaccine designs are currently undergoing laboratory testing at Perth and the University of Washington, using immune cell samples taken from HIV-infected patients. Researchers expect to release initial results later this year, but they're already optimistic about the research technique, suggesting it could also be used for work on hepatitis C and other mutating viruses.