A new method for protein tertiary structure prediction

• Authors: Yi Shang;Rajkumar Bondugula;Dong Xu;Qingguo Wang
• Venue: CI '07 Proceedings of the Third IASTED International Conference on Computational Intelligence
• Year: 2007

In this paper, we propose a new method for protein structure prediction based on the mini-threading approach. It has two main components, relevant protein fragment retrieval and protein 3-D structure construction. First, we develop new computational techniques to identify useful fragments in Protein Data Bank (PDB) for a given query protein. Then, different from the angular restraints used in most existing mini-threading methods, we formulate spatial restraints derived from the alignments between a query sequence and its fragment hits of known structures in Cartesian coordinates. We include additional information for the scoring function, such as pair-wise contact potential and apply classical multidimensional scaling method for 3-D structure construction. The structure is further refined by local optimization algorithms. The new method represents a novel framework that can utilize the information in the known protein structure database more effectively and efficiently than existing methods.