Department of Biophysics, Johns Hopkins University, 3400 N. Charles Street, Baltimore, Maryland 21218, USA.
Laboratory of Computational Biology, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892-5690, USA.
J Chem Phys. 2018 Aug 21;149(7):072321. doi: 10.1063/1.5027413.
We present the reservoir pH replica exchange (R-pH-REM) method for constant pH simulations. The R-pH-REM method consists of a two-step procedure; the first step involves generation of one or more reservoirs of conformations. Each reservoir is obtained from a standard or enhanced molecular dynamics simulation with a constrained (fixed) protonation state. In the second step, fixed charge constraints are relaxed, as the structures from one or more reservoirs are periodically injected into a constant pH or a pH-replica exchange (pH-REM) simulation. The benefit of this two-step process is that the computationally intensive part of conformational search can be decoupled from constant pH simulations, and various techniques for enhanced conformational sampling can be applied without the need to integrate such techniques into the pH-REM framework. Simulations on blocked Lys, KK, and KAAE peptides were used to demonstrate an agreement between pH-REM and R-pH-REM simulations. While the reservoir simulations are not needed for these small test systems, the real need arises in cases when ionizable molecules can sample two or more conformations separated by a large energy barrier, such that adequate sampling is not achieved on a time scale of standard constant pH simulations. Such problems might be encountered in protein systems that exploit conformational transitions for function. A hypothetical case is studied, a small molecule with a large torsional barrier; while results of pH-REM simulations depend on the starting structure, R-pH-REM calculations on this model system are in excellent agreement with a theoretical model.
我们提出了用于恒 pH 模拟的储层 pH 复制交换 (R-pH-REM) 方法。R-pH-REM 方法由两步程序组成；第一步涉及构象库的生成。每个储层都是通过标准或增强的分子动力学模拟得到的，其中质子化状态受到约束（固定）。在第二步中，当来自一个或多个储层的结构周期性地注入到恒 pH 或 pH 复制交换 (pH-REM) 模拟中时，固定电荷约束被放松。两步过程的好处是，构象搜索的计算密集部分可以与恒 pH 模拟解耦，并且可以应用各种增强构象采样的技术，而无需将这些技术集成到 pH-REM 框架中。使用受阻 Lys、KK 和 KAAE 肽的模拟来演示 pH-REM 和 R-pH-REM 模拟之间的一致性。虽然对于这些小测试系统不需要储层模拟，但在可离解分子可以采样两个或更多通过大能垒分离的构象的情况下，就需要储层模拟了，因此在标准恒 pH 模拟的时间尺度上无法实现足够的采样。这种问题可能会出现在利用构象转变来实现功能的蛋白质系统中。研究了一个假设的情况，即具有大扭转势垒的小分子；虽然 pH-REM 模拟的结果取决于起始结构，但对该模型系统的 R-pH-REM 计算与理论模型非常吻合。
