Oden Institute for Computational Engineering and Sciences, Department of Chemistry, The University of Texas at Austin, Austin, Texas 78712, USA; email:
Annu Rev Biophys. 2020 May 6;49:69-85. doi: 10.1146/annurev-biophys-121219-081528.
Recent advances in theory and algorithms for atomically detailed simulations open the way to the study of the kinetics of a wide range of molecular processes in biophysics. The theories propose a shift from the traditionally very long molecular dynamic trajectories, which are exact but may not be efficient in the study of kinetics, to the use of a large number of short trajectories. The short trajectories exploit a mapping to a mesh in coarse space and allow for efficient calculations of kinetics and thermodynamics. In this review, I focus on one theory: Milestoning is a theory and an algorithm that offers a hierarchical calculation of properties of interest, such as the free energy profile and the mean first passage time. Approximations to the true long-time dynamics can be computed efficiently and assessed at different steps of the investigation. The theory is discussed and illustrated using two biophysical examples: ion permeation through a phospholipid membrane and protein translocation through a channel.
近年来,原子细节模拟的理论和算法的进展为研究生物物理学中广泛的分子过程动力学开辟了道路。这些理论提出了从传统的非常长的分子动力学轨迹(虽然精确,但在研究动力学方面可能效率不高)向使用大量短轨迹的转变。短轨迹利用粗空间网格映射,并允许对动力学和热力学进行有效的计算。在这篇综述中,我专注于一种理论:里程碑理论是一种理论和算法,提供了对感兴趣的性质的分层计算,例如自由能分布和平均首次通过时间。可以高效地计算并在研究的不同步骤评估对真实长时间动力学的近似值。该理论使用两个生物物理实例进行了讨论和说明:离子通过磷脂膜的渗透和蛋白质通过通道的易位。