Laws of mechanics and function in proteins as evolved molecular machines

项目总结 蛋白质是执行生命所必需的化学反应的“机器”。与人为的不同 机器，蛋白质表现出独特的特征组合：相对于特定的 生物功能，以及对环境变化的适应性。这两个性质的重合 这表明，存在着支配进化机器的未知设计原则。然而，尽管 这一观察结果对我们理解进化、蛋白质和工程学具有明显的意义，这些 由于蛋白质内部原子运动的高维性和广泛性，设计原理仍然难以捉摸 与问题相关的长度和时间范围。目前，分子动力学模拟 一直是揭示蛋白质内部动态的最有用的工具，但考虑到挑战 在测量蛋白质经历某种生物过程的依赖于时间的内部运动时，这些 模拟常常不能被实验数据所证实。我们实验室新的实验和分析方法 能够以埃分辨率测量蛋白质内部的运动。这提供了一个机会：我将 进行经过大量实验验证的分子动力学模拟，模拟新的 实验。通过使用几个不同的力场执行这些模拟，并比较各种 量到它们的实验值，我将构建新的实验驱动的分子动力学“最好” 实践“使得模拟和实验意味着给定蛋白质的一致的性质。使用这些 优化的分子动力学方法以及降维等数据分析方法，即 将研究多种环境下的蛋白质动力学，包括电场刺激X射线结晶学(EFX)， 一种新技术，通过将电场施加到蛋白质晶体上，得到的结构是 用X射线结晶学和室温X射线结晶学(RTX)测量，一种新的方法 使用静态结晶学数据在室温下确定蛋白质构型的热系综。 这些将使我能够将数千个原子的原子尺度运动的测量减少到 对不同尺度上的协调运动的描述，以期揭示少数 决定蛋白质功能和变构的机械机制。这一数学分析不仅开启了 从微观上理解蛋白质功能的新的实验、计算和概念方法 结构信息；其改进的、基于证据的模拟方法可用于预测集体 当结晶学实验不存在或无法访问时的运动。总体而言，这项研究将继续发展 用于研究蛋白质力学和功能之间联系的新的定量工具，并实现 这些工具用来提取对这个看似高维、无序的低维、有序的描述 现象。

PROJECT SUMMARY Proteins are the "machines" that carry out the chemical reactions necessary for life. Unlike man-made machines, proteins exhibit a unique combination of features: high performance with respect to specific biological functions, and adaptability to changes in their environment. The coincidence of these two properties suggests that there exist yet-unknown design principles which govern evolved machines. However, despite the clear implications of this observation for our understanding of evolution, proteins, and engineering, these design principles remain elusive due to the high-dimensionality of internal protein atomic motions and wide range of length- and time-scales associated with the problem. At present, molecular dynamics simulations have been the most useful tools for shedding light on the internal dynamics of proteins, but given the challenge in measuring the time-dependent internal motions a protein undergoing some biological process, these simulations are often not confirmable by experimental data. New experiments and analysis methods in our lab are able to measure internal motions within proteins at Angstrom-resolution. This provides an opportunity: I will carry out substantial experimentally-verified molecular dynamics simulations which model the new experiments. By carrying out these simulations using several different force fields and comparing a variety of quantities to their experimental values, I will construct new experimentally-motivated molecular dynamics “best practices” such that simulations and experiment imply consistent properties of the given protein. Using these “optimized” molecular dynamics methods along with data analysis methods such as dimensional reduction, I will study protein dynamics in multiple contexts, including electric-field stimulated X-ray crystallography (EFX), a novel technique by which an electric field is applied to a protein crystal and the resulting structure is measured using X-ray crystallography, and room-temperature X-ray crystallography (RTX), a new way of determining thermal ensembles of protein configurations at room temperature using static crystallography data. These will enable me to reduce the measurements of atomic-scale motions of thousands of individual atoms to a description of coordinated motions on different scales with the expectation of revealing a small number of mechanical mechanisms dictating protein function and allostery. This mathematical analysis not only opens up new experimental, computational and conceptual methods for understanding protein function from microscopic structural information; its improved, evidence-based simulation methods can be used for predicting collective motions when crystallography experiments are nonexistent or inaccessible. Overall, this research will develop new quantitative tools for studying the connection between protein mechanics and function, and implement these tools to extract a low-dimensional, ordered description of this seemingly high-dimensional, disordered phenomenon.