Physics-based characterization of functionally relevant protein conformational dynamics

项目总结 随着结构生物学和超级计算技术的最新进展，全原子分子动力学 (MD)模拟技术作为研究蛋白质结构的重要工具，已经获得了发展的势头 动力学。然而，强力MD不能充分采样大多数功能相关的样本 生物分子过程，如大规模的蛋白质构象变化。已经有了各种方法 在过去的三十年里开发的，以解决阻碍MD在实时- 全球应用。自由能计算方法、增强的采样技术和路径查找 算法是描述这些方法中许多方法的伞形术语的例子。这个项目特别是 旨在使用、定制和微调最先进的增强采样和路径查找 解决重要的生物和生物医学问题的算法。这个项目的总体目标是 制定并采用可靠、实用的采样和分析方案，以研究功能相关性 从成纤维细胞生长因子到冠状病毒刺突蛋白的各种蛋白质的构象变化。我们的 建议的方法论框架特别利用(1)稳健的理论形式主义 根植于非平衡统计力学和微分几何；(2)系统特性增强 可针对手头具体问题调整的采样方案；(3)整合性和协同性 实验(特别是smFRET)和计算(特别是MD)技术的方法。一些人 建议在这里研究的系统包括质子偶联寡肽转运体、流感 血凝素，三磷酸腺苷结合转运体，冠状病毒刺突蛋白，大分子机械敏感通道 电导、膜插入酶YidC、5-羟色胺转运体和成纤维细胞生长因子(FGF)蛋白。 所有这些项目的共同主题是 这些蛋白质的功能。成功使用本项目中提出的方法将使 在分子水平上表征这些构象变化，并为 最先进的强化采样技术在真实世界生物学研究中的常规应用 有问题。

PROJECT SUMMARY With recent advances in structural biology and supercomputing technology, all-atom molecular dynamics (MD) simulation technique has gained momentum as a prominent tool for the study of protein structural dynamics. Brute-force MD, however, is not capable of adequately sampling most functionally relevant biomolecular processes such as large-scale protein conformational changes. Various approaches have been developed over the last three decades to address the “timescale gap” that hinders the use of MD in real- world applications. Free energy calculation methods, enhanced sampling techniques, and path-finding algorithms are examples of umbrella terms that describe many of these methods. This project specifically aims at employing, tailoring, and fine-tuning state-of-the-art enhanced sampling and path-finding algorithms to address important biological and biomedical questions. The overall aim of this project is to develop and employ robust and practical sampling and analysis protocols to study functionally relevant conformational changes of various proteins from fibroblast growth factor to coronavirus spike protein. Our proposed methodological framework specifically takes advantage of (1) robust theoretical formalisms rooted in nonequilibrium statistical mechanics and differential geometry; (2) system-specific enhanced sampling protocols that are tunable for the specific problem at hand; and (3) and integrative and synergistic approach to experimental (specifically smFRET) and computational (specifically MD) techniques. Some of the systems proposed to be studied here include proton-coupled oligopeptide transporters, influenza hemagglutinin, ATP-binding transporters, coronavirus spike proteins, mechanosensitive channel of large conductance, membrane insertase YidC, serotonin transporter, and fibroblast growth factor (FGF) protein. The common theme in all of these projects is the large-scale conformational changes involved in the function of these proteins. The successful use of the methodology proposed in this project will allow the characterization of these conformational changes at the molecular level and pave the groundwork for the routine application of state-of-the-art enhanced sampling techniques in the study of real world biological problems.