GADDLE Maps: General Algorithm for Discrete Object Deformations Based on Local Exchange Maps,Journal of Chemical Theory and Computation

当前位置： X-MOL 学术 › J. Chem. Theory Comput. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

GADDLE Maps: General Algorithm for Discrete Object Deformations Based on Local Exchange Maps
Journal of Chemical Theory and Computation ( IF 5.7 ) Pub Date : 2018-01-09 00:00:00 , DOI: 10.1021/acs.jctc.7b00861
J. Manuel Otero-Mato ₁ , Hadrián Montes-Campos ₁ , Martín Calvelo ₂ , Rebeca García-Fandiño ₃ , Luis J. Gallego ₁ , Ángel Piñeiro ₄ , Luis M. Varela ₁

Affiliation

A new method for switching between structures consisting of equivalent discrete and flexible objects with different particle representation and object configuration, including different resolution levels (number of particles per object), is reported. The method is fully general since it does not require any extra code nor additional database elements for new systems. It is based on a Monte Carlo sampling of the configurational space for each object type of the target system. The sampling is controlled by a Metropolis acceptance criterion of movements (translations, rotations, and relative deformations of the object configuration) that uses the generalized distance between the sets of particles at both representations. For Gaussian distributed distances, such a minimization procedure is equivalent to an optimization of χ² in a maximum likelihood method. This provides sound statistical support since the method leads to the most probable configuration of the system at each representation. The configurations obtained in this way are then used to create resolution exchange maps for each object type, which allows the extrapolation of the conversion to every object configuration throughout the whole system. As an example, the method is here tested with several molecular dynamics simulated systems (ionic liquids, cyclodextrins, cell-penetrating peptides, cyclic peptides, lipid bilayers, vesicles, heterogeneous organic molecules, DNA, and solvated proteins) for different resolution force fields (GROMOS, AMBER, OPLS, MARTINI) using GROMACS. In this context, the method can be applied to map structures described by any other pair of force fields, as well as to homogeneous and heterogeneous systems with many different molecules. The method is proved to be highly efficient since the time required for the mapping is practically independent of the number of molecules in the target system.

中文翻译：

GADDLE映射：基于局部交换图的离散对象变形的通用算法

报道了一种在由具有不同粒子表示和物体配置（包括不同分辨率级别（每个物体的粒子数量））的等效离散和柔性物体组成的结构之间进行切换的新方法。该方法是完全通用的，因为它不需要任何额外的代码，也不需要新系统的额外数据库元素。它基于目标系统每种对象类型的配置空间的蒙特卡洛采样。采样由Metropolis接受的运动准则（对象配置的平移，旋转和相对变形）控制，该准则使用两种表示形式的粒子集之间的广义距离。对于高斯分布的距离，这样的最小化过程等同于χ的优化²以最大似然法这提供了可靠的统计支持，因为该方法导致每个表示形式的系统最可能的配置。然后将以此方式获得的配置用于为每种对象类型创建分辨率交换图，这允许将转换外推到整个系统中的每个对象配置。例如，在此方法中，使用了几种分子动力学模拟系统（离子液体，环糊精，可穿透细胞的肽，环状肽，脂质双层，囊泡，异质有机分子，DNA和溶剂化的蛋白质）针对不同的分辨力场（使用GROMACS的GROMOS，琥珀色，OPLS，MARTINI）。在这种情况下，该方法可以应用于由任何其他成对力场描述的地图结构，以及具有许多不同分子的同质和异质系统。由于映射所需的时间实际上与目标系统中分子的数量无关，因此该方法被证明是高效的。

更新日期：2018-01-09

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11