Modeling Conformation of Protein Loops by Bayesian Network

• Published in: Life System Modeling and Intelligent Computing pp. 430 - 438
• Main Authors: Yang, Peng, Lü, Qiang, Yang, Lingyun, Wu, Jinzhen
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg
• Series: Lecture Notes in Computer Science
• Subjects: Bayesian network; bivariate von Mises distribution; Modeling protein loop
• ISBN: 3642156142; 9783642156144
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-642-15615-1_51

Modeling protein loops is important for understanding characteristics and functions for protein, but remains an unsolved problem of computational biology. By employing a general Bayesian network, this paper constructs a fully probabilistic continuous model of protein loops, refered to as LoopBN. Direct affection between amino acids and backbone torsion angles can be learned under the framework of LoopBN. The continuous torsion angle pair of the loops can be captured by bivariate von Mises distribution. Empirical tests are conducted to evaluate the performance of LoopBN based on 8 free modeling targets of CASP8. Experimental results show that LoopBN not only performs better than the state-of-the-art modeling method on the quality of loop sample set, but also helps de novo prediction of protein structure by providing better sample set for loop refinement.