Multiscale Theory and Simulation of Chemical Systems

科学中的许多重要问题都涉及作用于不同长度和时间尺度的现象。例如，在生物学中，配体与蛋白质的结合（原子事件）可以触发二级和三级结构的变化，然后影响蛋白质复合物的形成（介观事件），然后影响细胞中的生物活性（宏观事件）。这项建议旨在利用和发展多尺度建模方法，可用于回答涉及不同尺度的问题。它分为三个主题 *。1.利用粗粒或全原子分子动力学技术结合电荷迁移算法，研究气相蛋白质复合物离子的解离。 概念验证模拟将探索使用含有基本位点的系链来隔离复合物中的电荷，并在预定义的位置创建量身定制的排斥力，目标是产生位点定向解离力。 以预定方式解离蛋白质复合物的能力将极大地拓宽可用于通过质谱法分析蛋白质复合物结构的实验工具。 这些方法只需要对蛋白质进行常用的化学修饰。 有了这样的方法，质谱学家将能够解开蛋白质复合物中的相互作用，并将其转化回细胞中的生物功能。2.纳米粒子对脂质囊泡的影响将使用混合模拟方法进行研究，该方法以原子方式处理系统的某些部分，并使用粗粒度力场处理其他部分。 混合方法以物理上一致的方式处理这两个区域，同时允许在重要区域（原子区域）中描述详细的结构甚至化学效应。 由于对纳米颗粒毒性的健康担忧，人们对这个问题非常感兴趣。3.我们将从两个方面探讨一种在粗粒度模拟中校正动态时间尺度的方法。 首先，通过考虑一个完全用耗散粒子动力学描述的系统，其中所需的波动力由原子分子动力学确定。 第二，通过考虑一种混合方法，将分子动力学处理的原子区域与耗散粒子动力学处理的介观区域耦合起来。 在这两种情况下，结果都是由原子信息提供的系统的粗粒度描述，以实现自一致性。 具有精确动力学的粗粒度方法可以彻底改变各种领域的模拟应用，包括材料科学，生物学，化学和物理学。

Many important questions in science involve phenomena acting upon varying length and time scales. For example, in biology, the binding of a ligand to a protein (atomistic event) can trigger changes in secondary and tertiary structure which then affect protein complex formation (mesoscopic event) which then affects biological activity in a cell (macroscopic event). This proposal seeks to utilize and develop multiscale modelling methods which can be used to answer questions involving disparate scales. This is broken down into three main themes.******1. Using coarse-grained or all-atom molecular dynamics techniques coupled with charge migration algorithms, the dissociation of gas phase protein complex ions will be studied.  Proof of concept simulations will explore the use of basic-site containing tethers to sequester charge in a complex, and create repulsive forces tailored at predefined positions, with the goal of producing site-directed dissociation forces.  The ability to dissociate a protein complex in a predefined manner will broaden immensely the experimental tools available for analyzing protein complex structure by mass spectrometry.  These methods require only commonly used chemical modifications of proteins.  With such methods, mass spectrometrists will be able to unlock the interactions in protein complexes and translate this back into biological function in cells.******2. The effect of nanoparticles on lipid vesicles will be investigated using a hybrid simulation methodology that treats some parts of a system atomistically, and other parts with a coarse-grained force field.  The hybrid method treats the two regions in a physically consistent way, while at the same time allowing detailed structural and even chemical effects to be described in areas of importance (the atomistic region).  There is intense interest in this problem due to health concerns over nanoparticle toxicity.******3. A method for correcting dynamical timescales in coarse-grained simulations will be explored in two ways.  First, by considering a system described completely with dissipative particle dynamics in which the necessary fluctuating forces are determined from atomistic molecular dynamics.  Second, by considering a hybrid method coupling an atomistic region treated with molecular dynamics with a mesoscopic region treated with dissipation particle dynamics.  In both cases, the result will be a coarse-grained description of the system fed by atomistic information for self-consistency.  A coarse-grained method with accurate dynamics can revolutionize simulation applications in a wide variety of fields, including materials science, biology, chemistry, and physics.