College of Liberal Arts & Sciences

RNA-RBP interaction prediction

Thursday, June 5, 2014

June 5, Thu 2014
1:00 pm, MRB 200

Dr. Shiyong Liu

Department of Physics, Huazhong University of Science & Technology

RNA-RBP interaction prediction

RNA and RNA binding protein (RBP) form specific RNA- protein complexes, which play a key role in many life processes. Recently, hundreds of new RBPs have been discovered experimentally. More RBPs are expected to be found in the near future. Currently, the number of RNA-protein interactions confirmed is far less than the actual the number of RNA- protein interactions in cell. Thus, RNA-protein interaction prediction is necessary.

RNA can be folded into a variety of three-dimensional structures, thus forming RNA-protein complexes with RBP, which is as complex as protein-protein interactions. Thus, RNA-protein docking like protein-protein docking remains a challenge. Based on a previous scoring function DECK (BMC Bioinformatics, 2011, 280), we developed a scoring function DECK-RP for protein-RNA, with a high success rate (Sci Rep, 2013,1887). At the same time, we compiled a dataset with the experimental binding energy (Protein Sci, 2013, 1808), which will be used in the scoring function optimization.

Furthermore, based on various features from protein sequence including physicochemical properties and evolutionary information, a computational method RBPPR-seq was developed based on Support Vector Machine (SVM) with the best MCC value 0.79, accuracy 0.89 and the area under the ROC curve (AUC) 0.96 for predicting RNA-binding proteins.

Reference

Huang, Y., S. Liu, D. Guo, L. Li and Y. Xiao (2013). "A novel protocol for three-dimensional structure prediction of RNA-protein complexes." Sci Rep 3: 1887.

Liu, S. and I. A. Vakser (2011). "DECK: Distance and environment-dependent, coarse-grained, knowledge-based potentials for protein-protein docking." BMC Bioinformatics 12: 280.

Yang, X., H. Li, Y. Huang and S. Liu (2013). "The dataset for protein-RNA binding affinity." Protein Sci 22: 1808.

Xiaoli Zhang and Shiyong Liu (2014) Another SVM-based predictor for RNA-binding protein prediction to be presented in ICSB poster session.



One of 34 U.S. public institutions in the prestigious Association of American Universities
44 nationally ranked graduate programs.
—U.S. News & World Report
Top 50 nationwide for size of library collection.
—ALA
23rd nationwide for service to veterans —"Best for Vets," Military Times
KU Today