Jha Shantenu, Coveney Peter V, Laughton Charles A
Centre for Computational Science, University College London, London WC1H 0AJ, United Kingdom.
J Comput Chem. 2005 Nov 30;26(15):1617-27. doi: 10.1002/jcc.20300.
Important questions exist regarding the quality of force fields used in molecular dynamics (MD) simulations and their interoperable use with other available MD implementations. NAMD is one of the most efficient and scalable parallel molecular dynamics codes for large-scale biomolecular simulations in the open source domain. It is the aim of this article to analyze and compare the dynamics of a benchmark DNA dodecamer d(CTTTTGCAAAAG)2 system, including its binding to a specific drug molecule arising from the use of various simulation protocols in NAMD using Amber98, with the dynamics arising from simulations of the same dodecamer using Amber98 in the AMBER package, one of the most well-established simulation codes for nucleic acids. Based upon a set of validation benchmarks, the details of which are discussed, we find that nucleic acid simulations using NAMD give meaningful results and that the essential features of the resulting dynamics are similar to those arising from the AMBER package. This sets the stage for reliable large-scale simulations of nucleic acids using NAMD.
关于分子动力学(MD)模拟中使用的力场质量以及它们与其他可用MD实现的可互操作使用,存在一些重要问题。NAMD是开源领域中用于大规模生物分子模拟的最有效且可扩展的并行分子动力学代码之一。本文旨在分析和比较一个基准DNA十二聚体d(CTTTTGCAAAAG)2系统的动力学,包括其与使用Amber98在NAMD中通过各种模拟协议产生的特定药物分子的结合,以及使用Amber98在AMBER软件包(核酸领域最成熟的模拟代码之一)中对同一十二聚体进行模拟所产生的动力学。基于一组验证基准(其详细信息将进行讨论),我们发现使用NAMD进行的核酸模拟能给出有意义的结果,并且所得动力学的基本特征与AMBER软件包产生的特征相似。这为使用NAMD进行可靠的核酸大规模模拟奠定了基础。