Skip to Navigation Skip to Page Content

MU receives high marks in international protein structure prediction competition

Illustrated are four high-quality protein structure predictions submitted by Mizzou Engineering Professor Jianlin Cheng to the Institute of Health's Critical Assessment of Techniques for Protein Structure Prediction (CASP8). The orange experimental structures and the green predicted structure models are superimposed, and are almost identical.

The University of Missouri has raised its stature in the international protein modeling community with its remarkable success in the Eighth Critical Assessment of Techniques for Protein Structure Prediction (CASP8), sponsored by the National Institute of Health.

Results were released at the CASP8 conference held in Cagliari, on the island of Sardinia, Italy, in early December. The top four to five research teams in each assessment category were invited to do presentations at the conference. Based on the success of the software they developed, MU Computer Science Professors Dong Xu and Jianlin Cheng were invited to speak.

From May through August, CASP organizers sent protein sequences to research groups around the world, two or three proteins per working day, for a total of 128 proteins. Participating teams were required return predictions within three days, and independent experts evaluated submission.

Predictions were ranked in a number of categories, most notably, template-based and template-free protein modeling. Template-based modeling uses solved protein structures within the Protein Data Bank as templates to predict structure. Template-free modeling predicts protein structure without using solved protein structures, representing one of the most challenging problems in bioinformatics.

The success of new bioinformatic software used by Computer Science Department Chair Xu’s team of three labs, MUFold earned him an invitation to speak on template-free prediction.

Xu’s team members include Yi Shang, a professor in computer science, and Ioan Kosztin, an associate professor of physics. Bogdan Barz, a graduate student in physics; Zhiquan He and Qingguo Wang, graduate students in computer science; and Jingfen Zhang, a post-doc in medical sciences also members of the prediction teams.

“Protein structure prediction is a very important yet highly challenging problem in bioinformatics.” Xu said. “I feel fortunate to have two other MU labs with complementary expertise working closely together on this challenging topic. We developed several innovative methods to tackle this problem. I am glad that our collaboration paid off at CASP.”

Cheng methods with software MULTICOM, earned his team top honors in both major categories and he was invited to give talks in both. He also was selected to speak on model quality assessment and protein disordered region prediction.

“Once a protein is generated, it is important to evaluate the quality of the model for biologists to use it appropriately. I am pleased to see MULTICOM also excelled in this important category,” said Cheng whose team included graduate students Zheng Wang from computer science, and Allison Tegge from informatics.

Prediction of disordered regions-regions of a protein that do not adopt a fixed structure-is important for experimental structure determination by X-ray crystallography and Nuclear Magnetic Resonance (NMR), Cheng explained. Many disordered regions have important functions such as protein binding, he said.

In addition to the success in these four categories, MULTICOM was also ranked at the top in protein residue-residue contact prediction and protein domain boundary prediction.

“MULTICOM can be practically applied to many protein modeling problems in life science research,” said Cheng. “I look forward to collaborating with life scientists at MU and others from around the world to model protein structures of interest to their research.”

“The success of MU researchers at CASP further increased the visibility of MU in the field of bioinformatics.” Xu said. “It will help our graduate programs to attract more outstanding students, and it laid foundation for more research funding.”