Collaborative Research: CRIF: CRF: Extensible Cyberenvironments for Empirical and Semiempirical Hamiltonian Parameter Optimization and Dissemination

马里兰大学巴尔的摩分校的Alexander D. MacKerrell、伊利诺伊大学厄巴纳-香槟分校的Sudhakar V. Pamidighantam、佛罗里达大学的Adrian E. Roitberg以及肯塔基大学的John W.D. Connolly和Michael Sheetz得到了美国国家科学基金会化学部网络基础设施和研究设施项目的支持。该项目将创建网络环境，以自动化经典分子力学（MM）和半经验（SE）哈密顿量的参数化过程，并允许开发参数的广泛传播。项目目标包括一个可扩展的网络环境，用于i)新型哈密顿量的快速和系统参数化，ii)当前可用模型的系统扩展，通过网络环境提供i)和ii)的结果参数集。集成环境将包括一个用于参数化过程的实验和量子力学参考数据数据库，以及用于数据采集、量子力学参考数据生成自动化和参数优化过程自动化的计算资源。GridChem计算化学网格将作为量子化学和分子力学服务的计算网络基础设施，并将包括创建和执行特定参数化方案的工作流程。工作流管理工具专门用于生成质量管理参考数据、监控参数优化和分析，并将包括允许专家干预参数化过程的接口。许多现有流行的MM和SE哈密顿量将被整合，其中包括生物，有机和无机物种的广泛参数将可直接使用或进一步优化。该基础设施在数据源和能量函数方面是可扩展的，允许其适用于任何参数化方案。该项目将通过简化参数优化任务，使用理论方法更准确地描述各种材料、药理学和生物系统的静态和动态特性。这将允许为各种分子系统生成高质量的参数。建模精度的提高以及可获得化学系统的范围的扩大将使化学、纳米技术、医学和生物学等领域受益。此外，用于工程领域的分析模型，如结构力学和流体动力学，将成为分子水平处理的途径。

Alexander D. MacKerrell of the University of Maryland at Baltimore, Sudhakar V. Pamidighantam of the University of Illinois at Urbana-Champaign, Adrian E. Roitberg of the University of Florida, and John W.D. Connolly and Michael Sheetz of the University of Kentucky are supported by the NSF Division of Chemistry under the Cyberinfrastructure and Research Facilities Program. This project will create Cyberenvironments to automate the process of parameterization for classical molecular mechanics (MM) and semi-empirical (SE) Hamiltonians and allow for wide dissemination of the developed parameters. Project goals include an extensible cyberenvironment for i) the rapid and systematic parameterization of novel Hamiltonians and ii) the systematic extension of currently available models, with the resulting parameters sets from both i) and ii) to be made available via the cyberenvironment. The integrated environment will include a database of experimental and quantum mechanical reference data to be used in the parameterization process along with computational resources for data acquisition, automatization of QM reference data generation and automatization of parameter optimization processes. GridChem Computational Chemistry Grid will serve as the computational cyberinfrastructure for Quantum Chemical and Molecular Mechanics services and will include workflows from which specific parameterization schemes will be created and executed. Workflow management tools specific to generating QM reference data, monitoring parameter optimization and analysis will be implemented and will include interfaces allowing for expert intervention in the parameterization process. Many existing popular MM and SE Hamiltonians will be integrated from which a wide range of parameters encompassing biological, organic and inorganic species will be accessible for direct use or further optimization. The infrastructure will be extensible in terms of data sources and energy functions allowing for its applicability in any parameterization scheme.This project will provide for more accurate descriptions of the static and dynamic properties of a wide range of material, pharmacological and biological systems using theoretical methods by simplifying the task of parameter optimization. This will allow for the generation of high quality parameters for a wide variety of molecular systems. Improvements in the accuracy of modeling as well as the range of accessible chemical systems will benefit such fields as chemistry, nanotechnology, medicine and biology, among others. In addition, analytical models used in engineering fields such as structural mechanics and fluid dynamics will become accessible to molecular level treatments.