The source code, available as a free download from Microsoft's CodePlex Web site, is designed to use the software giant's machine-learning technology to sort through thousands of human immunodeficiency virus (HIV) strains. Researchers hope to use the technology to identify genetic patterns that could help them train an infected person's immune system to combat the virus.
AIDs researchers will have the choice of either downloading the four tools and using their preconfigured format or using the source code to develop their own applications.
One tool, PhyloD, seeks correlations between a patient's human leukocyte antigen (HLA)--a key component of the immune system--and the virus. A second tool, Epitope Prediction, is designed for people with any type of HLA and aims to scan proteins for the part of the antigen that elicits an immune response, or epitope.
The HLA Assignment tool, meanwhile, aims to improve the accuracy in finding epitopes, while the HLA Completion tool is designed to provide greater granular detail about a person's genetic makeup by addressing the hierarchy of his or her immune system's HLA types.
The HLA Completion tool was released Wednesday.
Microsoft began applying some of its technology to AIDS research in 2005, after it discovered its machine-learning technology could be used for such purposes. The research has included the efforts of roughly a dozen Microsoft researchers, who worked with doctors and scientists in Microsoft labs.
"We apply technology to some of the world's toughest technical and societal challenges," David Heckerman, lead researcher of Microsoft's Machine Learning and Applied Statistics Group, said in a statement. "And with 10,000 people per day dying of AIDS, this world health crisis is certainly one of those challenges."
