NIH grants MU researchers $1.18 million to develop protein structure prediction system

The National Institute of General Medical Sciences (NIGMS) of the National Institutes of Health (NIH) recently granted University of Missouri researchers $1.18 million to further develop and improve their protein prediction software system called MULTICOM. Using this system to understand protein structures in diseases can help scientists design drugs for a variety of diseases.

"Proteins are fundamental molecules that are involved in many basic and important functions," said Jianlin Cheng, assistant professor of computer science in the MU College of Engineering. "The structure of a protein determines the function. When a protein folds abnormally, dysfunctions occur, which lead to diseases, such as Parkinson's disease, Alzheimer's and cancer. By converting protein sequences into predictable structures, scientists can view how a structure folds and thus predict how drugs will affect the protein. This information is useful when designing drugs and cures."

The current "gold standard" for protein structure determination is X-ray crystallography and nuclear magnetic resonance (NMR) spectroscopy techniques. These experimental techniques can be time consuming and expensive, Cheng said. With this grant, Cheng would attempt to develop fast and cheap computational technology to accurately predict protein structure from protein sequence.

"A lot of advances have been made since this field was developed," Cheng said. "We are able to predict high-accuracy structures for about 50 percent of proteins, but we still have a lot of room for improvement."

Every two years, scientists who are developing protein-prediction technology test their programs in the Critical Assessment of Techniques for Protein Structure Prediction (CASP), which is considered by scientists to be the "Olympics of protein prediction." During the competition, CASP organizers send protein sequences to research groups around the world, two or three proteins per working day, for a total of more than 100 proteins. Researchers must return their predictions within three days. Independent experts evaluate the submissions. Teams that score the highest are invited to give presentations at the CASP conference.

In 2008, the MU team ranked at the top in several categories at CASP8. The team is participating in CASP9 this summer from May to July.