Frank Aaron T, Andricioaei Ioan
Department of Chemistry, The University of California, Irvine , 4212 Natural Sciences 1, Irvine, California 92697, United States.
J Phys Chem B. 2016 Aug 25;120(33):8600-5. doi: 10.1021/acs.jpcb.6b02654. Epub 2016 Jun 6.
Enhanced sampling techniques are used to increase the frequency of "rare events" during computer simulations of complex molecules. Although methods exist that allow accurate thermodynamics to be recovered from enhanced simulations, recovering kinetics proves to be more challenging. Here we present an extrapolation approach that allows reliable kinetics to be recovered from potential-scaled MD simulations. The approach, based on Kramers' rate theory, is simple and computationally efficient, and allows kinetics to be recovered without defining reaction coordinates. To test our approach, we use it to determine the kinetics of barrier crossing between two metastable states on the 2D-Müller potential and the C7eq to αR transition in alanine dipeptide. The mean first passage time estimates obtained are in excellent agreement with reference values obtained from direct simulations on the unscaled potentials performed over times that are orders of magnitude longer.
在复杂分子的计算机模拟中，增强采样技术用于增加“罕见事件”的发生频率。尽管存在一些方法可从增强模拟中准确恢复热力学性质，但事实证明恢复动力学更具挑战性。在此，我们提出一种外推方法，该方法能够从势能缩放的分子动力学（MD）模拟中可靠地恢复动力学。该方法基于克莱默斯速率理论，简单且计算效率高，无需定义反应坐标即可恢复动力学。为了测试我们的方法，我们用它来确定二维穆勒势上两个亚稳态之间的势垒穿越动力学以及丙氨酸二肽中从C7eq到αR的转变。所获得的平均首次通过时间估计值与通过对未缩放势能进行直接模拟得到的参考值高度吻合，而直接模拟所需的时间要长得多，达到几个数量级。
