The TINKER Modeling Software for Biomoledular Simulation

Tinker分子建模包是一套模块化的、可扩展的计算生物学 旨在作为经验势场(“力场”)和算法开发的试验台的程序。Tinker以可执行代码和源代码的形式免费提供，并在世界各地的学术和工业实验室中得到了广泛的使用。该程序包的一个重要功能是它能够使用许多可供选择的当代力场参数集执行能量计算。目前的项目打算将这种比较能力扩展到用于生物分子模拟的新兴可极化能量模型。进一步开发Tinker的力场浏览器图形界面将使 可视化静电效应，如极化，以及远程监控修补程序计算。分布式内存并行化、先进的Ewald求和方法和其他代码优化将允许在商用Beowulf集群和类似的计算机硬件上高效使用Tinker。 力场是源于化学物理经典原理的大量原子级生物分子建模技术的基石。因此，“分子力学”软件在各种研究和教育环境中都有用处。该项目的一个主要目标是产生一个“下一代”能量模型，它将为配体结合的热力学和分子内生物聚合物相互作用提供常规的0.5kcal/mol或更高的“化学精度”。尤其是基于结构的药物设计和高分辨率同源建模将受益于由此带来的准确性改进。这个 阿米巴可极化原子多极模型将扩展到核酸，该模型提供了对潜在的永久静电和对分子环境的响应的灵活和准确的描述。有记录的源代码和参数的现成可用将使新的极化势能够合并到其他广泛使用的建模套件中。应用将集中在生物重要性体系中绝对结合常数的计算、蛋白质环建模、侧链旋转体预测和结构优化。由于极化被广泛认为在离子通过膜蛋白通道的传输、离子迁移率和DNA沟槽中的浓度等问题中起关键作用，因此还将研究含离子体系的结构和能量学。

The TINKER Molecular Modeling package is a modular, extensible set of computational biology programs intended to serve as a test-bed for empirical potentials ("force fields") and algorithm development. TINKER is freely available in both executable and source code form, and has seen extensive use in both academic and industrial labs worldwide. An important capability of the package is its ability to perform energy calculations with a number of alternative present-generation force field parameter sets. The current project intends to extend this comparative capability to emerging polarizable energy models for biomolecular simulation. Further development of the Force Field Explorer graphical interface to TINKER will enable visualization of electrostatic effects, such as polarization, and remote monitoring of TINKER calculations. Distributed memory parallelization, advanced Ewald summation methods, and other code optimizations will allow efficient use of TINKER on commodity Beowulf clusters and similar computer hardware. Force fields are the cornerstone of a vast array of atomic-level biomolecular modeling techniques derived from classical principles of chemical physics. As such, "molecular mechanics" software sees use in a variety of research and educational settings. A primary goal of this project is to produce a "next-generation" energy model that will routinely provide "chemical accuracy" of 0.5 kcal/mol or better for the thermodynamics of ligand binding and for intramolecular biopolymer interactions. Structure-based drug design and highresolution homology modeling, in particular, will benefit from the resulting accuracy improvements. The AMOEBA polarizable atomic multipole model, which provides a flexible and accurate description of the underlying permanent electrostatics and response to the molecular environment, will be extended to nucleic acids. The ready availability of documented source code and parameters will enable the incorporation of the new polarizable potential into other widely-used modeling suites. Applications will focus on calculation of absolute binding constants in systems of biological importance, protein loop modeling, side chain rotamer prediction and structure refinement. The structure and energetics of ion-containing systems will also be investigated, since polarization is widely considered to play a key role in problems such as ion transport through membrane protein channels, ion mobility and concentration in DNA grooves.