State Key Laboratory of Microbial Metabolism, Department of Bioinformatics and Biostatistics, National Experimental Teaching Center for Life Sciences and Biotechnology, School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, China.
Department of Molecular Biology and Biochemistry, University of California, Irvine, CA 92697, USA.
Bioinformatics. 2019 Nov 1;35(22):4664-4670. doi: 10.1093/bioinformatics/btz298.
Protein residue interaction network has emerged as a useful strategy to understand the complex relationship between protein structures and functions and how functions are regulated. In a residue interaction network, every residue is used to define a network node, adding noises in network post-analysis and increasing computational burden. In addition, dynamical information is often necessary in deciphering biological functions.
We developed a robust and efficient protein residue interaction network method, termed dynamical important residue network, by combining both structural and dynamical information. A major departure from previous approaches is our attempt to identify important residues most important for functional regulation before a network is constructed, leading to a much simpler network with the important residues as its nodes. The important residues are identified by monitoring structural data from ensemble molecular dynamics simulations of proteins in different functional states. Our tests show that the new method performs well with overall higher sensitivity than existing approaches in identifying important residues and interactions in tested proteins, so it can be used in studies of protein functions to provide useful hypotheses in identifying key residues and interactions.
Supplementary data are available at Bioinformatics online.
蛋白质残基相互作用网络已成为理解蛋白质结构和功能之间复杂关系以及功能如何调节的有用策略。在残基相互作用网络中，每个残基都用于定义一个网络节点，这在网络后分析中会引入噪声并增加计算负担。此外，在破译生物功能时通常需要动态信息。
我们通过结合结构和动态信息，开发了一种强大而高效的蛋白质残基相互作用网络方法，称为动态重要残基网络。与以前的方法的一个主要区别是，我们试图在构建网络之前识别对功能调节最重要的重要残基，从而得到一个更简单的网络，其节点是重要残基。重要残基是通过监测来自不同功能状态下蛋白质的分子动力学模拟的集合结构数据来识别的。我们的测试表明，该新方法在识别测试蛋白质中的重要残基和相互作用方面表现良好，整体敏感性高于现有方法，因此可用于蛋白质功能研究，为识别关键残基和相互作用提供有用的假设。
补充数据可在“Bioinformatics”在线获取。
