CAREER - New Methods for Simulating Biomolecules of Several Microns in Length

分子功能的很大一部分涉及在广泛的长度范围内的运动，例如，从化学键的振动到超分子络合物的整体构象变化，再到宏观的肌肉收缩。这项研究的目标是引入几种新的计算方法来描述在任何期望的长度尺度上的运动，而不丢失原子计算的细节，这是一种前所未有的模拟能力。中心方法是子结构综合法，它将给定的结构视为以某种方式共同作用的子结构的集合。亚结构的选择是任意的，有时是非常自然的，例如超分子复合体中的结构域或亚单位。首先，通过求解特征值问题来确定每个子结构的振动模式。接下来，通过一组约束将不同的子结构连接在一起，以在子结构间界面强制几何兼容性。然后利用瑞利-里兹原理，利用一组低频子结构振型，计算出组合结构的振型。在计算上，这是一个比解决装配结构的全部特征值问题更可取的问题。这种新的方法将应用于F-肌动蛋白这一典型的微米丝状体系，以研究其任意长度的力学和动力学性质。这也将有助于解释测量分子弹性性质的实验。这种模拟大范围运动的新方法不仅将对分子动力学的研究产生影响，而且将对生物工程和化学领域做出重大贡献。此外，由于所提出的方法中的一些概念与机械工程领域的概念有关，因此它们作为描述复杂分子运动的方法的实现需要包括生物、物理和工程在内的多个学科的知识。因此，该项目的成功将是跨学科交流所带来的创造性思维和问题解决的一个很好的例子。受益者将不仅是直接参与该项目的受训人员，还将有更多具有不同背景的受众，他们对生命科学和工程学中的问题共同感兴趣，但角度非常不同。特别是，它将使本科生接触到跨学科研究的概念。最后，通过公开分发在这项工作中开发的计算机软件，将接触到更广泛的社区。

A substantial part of molecular functions involves motions in a wide range of length scales, e.g., from the vibrations of chemical bonds to global conformational changes of supermolecular complexes to macroscopic muscle contractions. The goal of this research is to introduce several new computational methods for describing the motions at any desired length scale without losing the details of atomic calculations, which is an unprecedented simulation capacity. The central method is substructure synthesis method that regards a given structure as an assemblage of substructures acting together in some ways. The choice of substructures is arbitrary, and sometimes quite natural, such as domains or subunits in supermolecular complexes. First, the vibrational modes for each substructure are determined by solving an eigenvalue problem. Next, various substructures are joined together by a set of constraints to enforce geometric compatibility at the inter-substructure interfaces. The modes for the assembled structure can then be computed by the Rayleigh-Ritz principle using a set of low-frequency substructure modes. Computationally, this represents a much more desirable problem than solving the full eigenvalue problem for the assembled structure. This new methods will be applied to F-actin, a typical filamentous system of several microns, to study its mechanical and dynamic properties at any length. This will also help interpret the experiments that measure molecular elastic properties. The new methods for simulating motions in a wide range of length scales will not only have an impact on the study of molecular dynamics, but also will significantly contribute to the fields of bioengineering and chemistry. Moreover, since some concepts in the proposed methods are related to those in the mechanical engineering field, their implementation as methods for describing motions of complex molecules requires knowledge of multiple disciplines including biology, physics, and engineering. The success of this project will therefore be an excellent example of creative thinking and problem solving benefited from interdisciplinary communication. The beneficiaries will not only be the trainees who are directly involved in the project, but also a larger audience with diverse background who are commonly interested in problems in life sciences and engineering, but from very different perspectives. Particularly, it will expose undergraduate students to concepts of interdisciplinary research. Finally, a broader community will be reached via public distribution of the computer software developed in this work.