Approaches to computing diffusion rates in proteins from transition path theory

项目描述（由申请人提供）：该项目的总体目标是使用并进一步发展过渡路径理论的数学框架(TPT； W. E. and E. Vanden-Eijnden, Annu。启。化学。2010(61):391-420)来量化蛋白质中小分子进入/退出和分子内扩散的速率。该项目基于全原子分子动力学，结合了单次扫描的自由能重建、零温度弦法的路径估计、提交函数的确定以及等提交函数划分面的里程碑计算。重点是使用提交器得出关于速率决定机制的推论，并使用等提交器叶子来提高里程碑计算的计算效率，以计算精确的速率。我们考虑了三种主要的渗透/蛋白质系统：(i) CO/肌红蛋白；O2/单体肌氨酸氧化酶（MSOX）；（iii） H2O/aquaporin-1。在每种情况下，速率量化将有助于评估假定的入口/出口入口和运输途径对总体动力学的相对贡献。到目前为止，还没有发表模拟方法来预测配体进入和分子内运输的速率。重要的是，TPT为计算用于统计描述反应轨迹的提交者函数提供了理论基础。本项目的关键智力贡献将展示提交者在上述重要蛋白质/配体系统中的计算，并使用它来识别速率限制机制，并提高计算精确速率的里程碑的效率和准确性。然而，尽管TPT使提交者的确定在概念上很简单，但迄今为止仍未解决如何最好地计算它，然后将其应用于大规模生物分子模拟的实践中。如果成功，预计这些方法将与实验动力学数据相比，允许对假定的运输途径和进入/退出门户进行有意义的评估。

DESCRIPTION (provided by applicant): The overall objective of this project is to use and further develop the mathematical framework of transition-path theory (TPT; W. E and E. Vanden-Eijnden, Annu. Rev. Phys. Chem. 2010 61:391-420) to quantify rates of small molecule entry/exit and intramolecular diffusion in proteins. Based on all-atom molecular dynamics, the project combines free-energy reconstruction using single sweep, pathway estimation using zero-temperature string method, determination of the committor function, and milestoning calculations with isocommittor dividing surfaces. The focus is on drawing inferences regarding rate-determining mechanisms using the committor and on using isocommittor foliation to enhance the computational efficiency of milestoning for computing exact rates. We consider three major penetrant/protein systems: (i) CO/myoglobin; (ii) O2/monomeric sarcosine oxidase (MSOX); and (iii) H2O/aquaporin-1. In each case, rate quantification will serve to evaluate the relative contribution of putative entry/exit portals and transport pathways to overall kinetics. To date, no simulation methods have been published which predict rates of ligand entry and intramolecular transport. Importantly, TPT provides the theoretical basis for computing the committor function which is used in a statistical description of reactive trajectories. The key intellectual contribution of this project will be demonstrating the computation of the committor in above important protein/ligand systems as well as using it to identify rate-limiting mechanisms and to enhance the efficiency and accuracy of milestoning for computing exact rates. However, although TPT makes committor determination conceptually straightforward, it remains so far unresolved how best to compute it and then apply it in practice in large-scale biomolecular simulations. If successful, it is anticipated that these approaches will allow for meaningful evaluation of putative transport pathways and entry/exit portals when compared to experimental kinetic data.