Scientists studying diseases will benefit from new protein interactions database

University of Michigan announces availability of its new Michigan Molecular Interactions (MiMI) database

January 24, 2007 - ANN ARBOR, MI – Research on disease processes will accelerate with a new multi-purpose protein database launched by the University of Michigan. This new tool will help biomedical scientists digest the enormous volumes of data being produced by modern biotechnology.

This database, called Michigan Molecular Interactions (MiMI) index, gathers data from multiple well-known protein interaction repositories and merges the information to create a new database using novel computational technology. Using MiMI, scientists can efficiently access information about protein processes and rapidly dig down to primary source data.

Biomedical scientists today often find valuable information in multiple data sources. When they visit these multiple websites and download data, they find that much of it is overlapping, and some is even contradictory. MiMI brings all these independent data together to form a complete picture of the protein interactions. Without MiMI this compilation would require considerable time and effort by a human expert.

“If you want to examine a single protein, you could visit ten independent websites and then synthesize the information,” says Hosagrahar V. Jagadish, Ph.D., professor of electrical engineering and computer science, College of Engineering, who led the development of the deep integration technology underlying the MiMI database. “If you want to select one target among a set of 200 proteins, you can do it in half a day using MiMI or 200 half days if you do it manually, the old way. With MiMI, the information from the multiple sites are available and are integrated for you into a single database.”

“Protein interactions underlie all of biochemistry and cellular regulation,” says David J. States, Ph.D., professor of human genetics and a senior scientist on the MiMI project. “Many proteins encoded by the human genome are poorly understood. Knowing what partners they interact with is a powerful tool in functional analysis.”

For example, researchers exploring the genetic causes of bipolar depression found linkages between two genes suspected of being involved in this disease. These same genes also play a role in apoptosis, or programmed cell death, which may be important in neural degeneration and hints at novel therapeutic strategies.

MiMI is hosted by the Michigan National Center for Integrative Biomedical Informatics (NCIBI), one of seven National Institutes of Health (NIH) “Roadmap” centers for biomedical computation. NCIBI at the University of Michigan is led by Brian D. Athey, Ph.D., associate professor of biomedical informatics, and associate director, Center for Computational Medicine and Biology (CCMB).

“MiMI will make biomedical researchers more productive,” says Athey. “It isn’t expected to solve medical mysteries by itself. But it will help researchers evaluate more hypotheses more quickly, and that should help us reach medical breakthroughs faster. It is a great new tool for our biomedical researchers!”

“The excellent multidisciplinary research environment at Michigan made it possible to put together the team needed to implement MiMI,” says Gilbert S. Omenn, M.D., Ph.D., a U-M professor of internal medicine, human genetics and public health, and director, CCMB. “This new database is a terrific example of how effective collaboration can enhance our research capabilities.”

MiMI is an ongoing research collaboration between the U-M Electrical Engineering and Computer Science Department, U-M Medical School, CCMB, and NCIBI. The National Science Foundation, NIH, Michigan Center for Biological Information, Pfizer, Howard Hughes Medical Institute, and Microsoft Corporation sponsor MiMI.

Written by Bruce Spiher

Contact Information

UMHS media relations

University of Michigan Health System

Contact via E-mail