AMBER force field consortium: a coherent biomolecular simulation platform

描述（由申请人提供）：分子模拟在生物化学和生物物理科学中发挥了重要作用。已经取得的进步使得可以对日益复杂的系统进行广泛的模拟。为了进一步推动这些进步，这组研究人员建议建立一个联盟来开发参数和模拟方法，作为分子模拟平台基础的一部分。拟议的联盟汇集了一群年轻的研究人员，他们都是加州大学旧金山分校科尔曼实验室的前成员，将“琥珀”力场的努力推向新的水平。这些研究人员在琥珀模拟包和力场参数的开发方面做出了重大贡献，并且在过去进行了重要的合作，这从作者之间的许多联合出版物中可以看出。该提案的一个重点是不仅为蛋白质、核酸和类药物分子开发通用、可靠和广泛适用的力场，而且通过彻底的测试和与其他可用方法和力场的比较来验证力场。目前，模型（极化、电荷模型、溶剂表示）的选择仍然是活跃的研究问题。该提案影响广泛，因为将研究多种方法，从下一代通用极化力场的开发到专门设计用于利用连续溶剂的模型。从组建的团队中可以看出，他们在蛋白质和核酸模拟方面拥有丰富的专业知识，这将有助于指导工作的进展。该联盟的一个主要目标是进一步加强密切合作，使想法能够更快地得到测试和审查。拟议的工作大致分为以下几个领域。 1) 将开发基于 Thole 偏振方案的完全偏振模型。这种力场有望改善生物分子与二价离子、药物和环境相互作用的表现； 2）将开发一种新的隐式溶剂模型，以纳入固定和动态溶剂以及溶质的极化率； 3）进一步开发加性和联合原子模型，以开发一种混合模型，该模型将关键的极化效应与典型的定点电荷模型的计算成本相当，并细化侧链参数以改进能量表面的表示； 4) 通用琥珀力场（GAFF）模型的开发将能够更准确地表示不同组类药物分子与生物分子相互作用的情况，这些生物分子以改进的加性和极化力场为代表； 5) 将通过与实验的直接比较来严格审查和严格评估模拟方法和相关参数。

DESCRIPTION (provided by applicant): Molecular simulations have played important roles in biochemical and biophysical sciences. Advances have been made that have allowed extensive simulations of increasingly complex systems. Furthering these advances, this group of investigators proposes to establish a consortium to develop parameters and simulation methodologies that are part of the foundation of molecular simulation platform. The proposed consortia brings together a group of young researchers, all former members of the Kollman lab at UCSF, to push the "Amber" force field efforts to the next level. These investigators have made significant contributions :o the development of both Amber simulation package and the force field parameters and have undertaken significant collaboration in the past which is evident by many joint publications among the authors. A key focus of this proposal is to not only develop general, reliable and widely applicable force fields for proteins, nucleic acids and drug-like molecules, but to validate the force fields via thorough testing and comparison to other available methods and force fields. At present, the choices to make in terms of the model (polarization, charge model, solvent representation) are still active research questions. This proposal is broad-reaching in that multiple approaches will be investigated, ranging from development of next generation general-purpose polarizable force field to models specifically designed to take advantage of continuum solvent. From the team assembled, considerable expertise in the simulation of proteins and nucleic acids is evident which will help guide the efforts forward. A key objective of the consortium is to further enhance the close collaboration that allows ideas to be tested and vetted much more quickly. The proposed work is broadly categorized in the following areas. 1) A fully polarizable model based on Thole's scheme of polarization will be developed. Such a force field is expected to improve the representation of biomolecules interacting with divalent ions, drugs and the environment; 2) A new implicit solvent model will be developed to incorporate both fixed and dynamic solvent and solute polarizability; 3) Further development of additive and united-atom models will be pursued to develop a hybrid model that incorporates the key polarization effect with a comparable computational cost to that of typical fixed point charge models and to refine the side chain parameters for an improved representation of the energetic surface; 4) Development of the general Amber force field (GAFF) model will allow more accurate representation of diverse sets of drug-like molecules interacting with biomolecules represented by improved additive and polarizable force fields; 5) The simulation methodology and the associated parameters will be vigorously scrutinized and critically assessed through direct comparisons with experiments.