Hamiltonian replica exchange combined with elastic network analysis to enhance global domain motions in atomistic molecular dynamics simulations
Hamiltonian replica exchange combined with elastic network analysis to enhance global domain motions in atomistic molecular dynamics simulations
复制标题
DOI：
10.1002/prot.24695
复制
发表时间：
2014-12-01
影响因子：
2.9
通讯作者：
Zacharias, Martin
中科院分区：
文献类型：
--
作者：
Ostermeir, Katja;Zacharias, Martin
蛋白质的粗粒度弹性网络模型（ENM）提供了蛋白质动力学和全局移动方向的低分辨率表示。一个哈密尔顿副本交换分子动力学（H-REMD）的方法已经开发，结合从ENM分析与原子显式溶剂MD模拟提取的信息。基于一组代表蛋白质刚性片段（质心）的中心，通过ENM分析构建了一个距离依赖的偏置电位，以促进和引导质心/结构域重排。在原力场的控制下，沿沿着具有一个参考副本的副本，将具有不同幅度的偏置电势添加到MD模拟的力场描述中。偏置电位的大小和形式在模拟期间基于平均采样构象进行调整，以在平衡后在每个副本中达到接近恒定的偏置。这允许在每个副本中的构象状态的规范采样。该方法的应用到一个两个结构域的糖蛋白130和蛋白质cyanovirin-N的部分表明显着增强的全球域运动和改进的构象采样相比，传统的MD模拟。Proteins 2014; 82：3410 - 3419. (c)2014年威利期刊公司
Coarse-grained elastic network models (ENM) of proteins offer a low-resolution representation of protein dynamics and directions of global mobility. A Hamiltonian-replica exchange molecular dynamics (H-REMD) approach has been developed that combines information extracted from an ENM analysis with atomistic explicit solvent MD simulations. Based on a set of centers representing rigid segments (centroids) of a protein, a distance-dependent biasing potential is constructed by means of an ENM analysis to promote and guide centroid/domain rearrangements. The biasing potentials are added with different magnitude to the force field description of the MD simulation along the replicas with one reference replica under the control of the original force field. The magnitude and the form of the biasing potentials are adapted during the simulation based on the average sampled conformation to reach a near constant biasing in each replica after equilibration. This allows for canonical sampling of conformational states in each replica. The application of the methodology to a two-domain segment of the glycoprotein 130 and to the protein cyanovirin-N indicates significantly enhanced global domain motions and improved conformational sampling compared with conventional MD simulations. Proteins 2014; 82:3410-3419. (c) 2014 Wiley Periodicals, Inc.