Microsoft Supports Human Genome Research

April 17, 2008

Breakthrough research with the ability to predict and prevent adverse drug responses within prescription medicine, and provide greater insight into the cause of neurodegenerative illnesses such as Parkinson’s disease and Alzheimer’s disease, are just two of six research projects Microsoft Research will support through its Computational Challenges of Genome Wide Association Studies (GWAS) program.

The supported projects were selected from 40 proposals submitted from 39 academic institutions worldwide. Microsoft Research’s goal with funding computational challenges associated with GWAS research is to aid researchers in the correlation of genetic patterns with patients’ responses to drugs, diseases, aging or the expression of genetic disorders. The ultimate goal is to enable lifesaving research through the use of technology.

“The Microsoft Research GWAS program provides crucial funding at the interface between information management and quality healthcare,” said Michael Kane, an assistant professor of Computer Technology at Purdue University and one of the researchers selected to receive support under the GWAS program. “Patient-specific genotyping to assure prescription drug safety and drug effectiveness is a major step toward the emergence and adoption of personalized medicine, and this support is key to facilitating that vision. Microsoft Corp. has recognized the important role information technology will play in the future of healthcare. Ultimately, this is about technology that helps to save lives.”

As part of the program, more than $850,000 was shared among six research institutions.

“When it comes to performing genetic analysis, researchers are often hampered by the data itself, whether it’s inconsistencies in format, the inability to visualize it, or sheer volume,” said Kristin Tolle, program manager for biomedical computing on the External Research & Programs (ER&P) team at Microsoft Research. “Through this program, Microsoft Research is encouraging the development of computer-science solutions to improve data access, standardization, visualization and tools to help scientists study the human genome.

“Of all the research areas Microsoft collaborates with, none may have as much real impact on global economies and the lives of ordinary people as healthcare,” Tolle said.

Enabling this research is part of ER&P’s broader efforts to move research in new directions across nearly every field of computer science, engineering and general science. Through this and other regional and global programs, Microsoft Research is partnering with researchers and scientists to address some of the toughest, most urgent scientific and societal challenges — such as those in healthcare — through enabling technologies that can provide real solutions.

Other examples of Microsoft’s support of healthcare-related research include the program Cell Phones as a Platform for Healthcare, which sought novel healthcare solutions that are accessible, affordable and relevant for smart mobile phones; and the Intelligent Systems for Assisted Cognition Awards, which provided funding for technologies to assist people living with disorders such as autism and Alzheimer’s disease. In the past six months alone, Microsoft External Research & Programs has awarded more than $2.5 million to academic researchers and scientists in support of healthcare research.

Details of the six winners and a description of their projects are as follows:

• Purdue University, Michael Kane and John Springer: “PGRx: An Interactive Software System for Integrating Clinical Genotyping With Prescription Drug Safety Assurance.” According to a survey published in the Journal of American Pharmacists Association in 2001, drug-related mortality and morbidity was estimated at approximately 10 percent of all healthcare costs in the U.S. It is estimated that adverse drug reactions are the cause of more than 200,000 deaths each year. Kane and Springer will work to develop a software and data management system to predict and prevent adverse drug responses, and provide in-depth training for physicians and pharmacists to better understand the link between genes, drug metabolism and the risk of adverse drug responses within prescription medicine.

• Translational Genomics Research Institute, John Pearson: “A Universal Data Format for Genotype Microarrays.” Combining data generated in genome research is problematic due to the variety of software platforms in use today. Pearson will work to create a universal data format that would accommodate multiple vendor platforms into a single file and software library. The software library would allow for open use by the research community and commercial proprietary use by platform vendors.

• National Institutes of Health; Johns Hopkins Hospital, Bryan Traynor: “Genome Wide Association Study of Amyotrophic Lateral Sclerosis in Finland.” Amyotrophic Lateral Sclerosis (ALS) is a rapidly progressive, fatal disorder. The overall purpose of this project is to discover the genes that are relevant to development of ALS by studying 489 Finnish ALS cases. A more comprehensive understanding of ALS biology also may provide insight into the pathogenesis of other neurodegenerative diseases such as Parkinson’s disease and Alzheimer’s disease.

• Department of Bioengineering, University of California, San Diego; Division of Computer Science, University of California, Berkeley; Trey Ideker and Richard Karp: “Pathway-Based Association: A New Paradigm for Genome Wide Association Studies.” Ideker and Karp propose to help explain the associations captured by GWAS in terms of known gene and protein interactions by developing computational tools that help explain linkages between signaling, regulatory and metabolic pathways to the genes that are associated with a disorder. If successful, this research could have a positive impact on a broad range of genomic studies.

• Columbia University, George Hripcsak: “Phenotypic Pipeline for Genome-wide Association Studies.” Large-scale studies involving many subjects, or even smaller studies in which subjects are selected from a larger population, will require innovative means to extract a reliable, useful phenotype from electronic health records data. Hripcsak proposes to develop advanced informatics methods to convert raw health records data into usable research information.

• University of the Republic of Uruguay; Pasteur Institute at Montevideo, Raul Ruggia and Hugo Naya: “Data Quality Management for Model Improvement in GWAS.” This project addresses the problems of building a data-quality management environment for the biological area, which would enable the user to define and evaluate biological-oriented data-quality properties over specific data sources. The biological-oriented properties would be defined in terms of the basic ones, and the environment would use the existing techniques that manage basic quality properties. The main expected outcomes will consist of biological-oriented data-quality properties and a prototyped environment to manage and evaluate these quality properties on biological databases.