Department of Chemistry and Applied Biosciences, ETH Zurich, and Facoltà di Informatica, Istituto di Scienze Computazionali, Università della Svizzera Italiana, Via G. Buffi 13, 6900, Lugano, Switzerland.
J Comput Chem. 2018 Sep 30;39(25):2126-2132. doi: 10.1002/jcc.25386. Epub 2018 Oct 11.
We develop a Gromacs implementation of dimer metadynamics (DM) (JCTC 13, 425 [2017]) for enhanced sampling through artificial delocalization effects. This implementation is based entirely on a Plumed collective variable developed for this purpose, the fine tuning of Gromacs input parameters, modified forcefields and custom nonbonded interactions. We demonstrate this implementation on alanine dipeptide in vacuum and in water, and on the 12-residue alanine polypeptide in water and compare the results with a standard multiple-replica technique such as parallel tempering. In all the considered cases, this comparison is consistent and the results with DM are smoother and require shorter simulations, thus proving the consistency and effectiveness of this Gromacs implementation. © 2018 Wiley Periodicals, Inc.
我们开发了一个用于增强采样的人工离域效应的 Gromacs 二聚体元动力学（DM）（JCTC 13, 425 [2017]）实现。这种实现完全基于为此目的开发的 Plumed 集体变量，微调了 Gromacs 的输入参数、修改了力场和自定义非键相互作用。我们在真空和水中对丙氨酸二肽以及在水中对 12 残基丙氨酸多肽进行了此实现的演示，并将结果与标准的多副本技术（如平行温度）进行了比较。在所有考虑的情况下，这种比较是一致的，DM 的结果更平滑，需要的模拟时间更短，从而证明了这种 Gromacs 实现的一致性和有效性。© 2018 Wiley Periodicals, Inc.
