CAREER: CDS&E: Theory, Models and Computer Simulation of Molecular Recognition Processes

纽约城市大学布鲁克林学院的Emilio Gallicchio教授获得了化学系化学理论、模型、计算方法和生命过程化学项目的奖励，以开发理论模型、计算算法和模拟软件来研究分子识别过程。分子识别——分子识别和特异性结合其他分子的能力——是所有物理化学过程的一个基本方面，在细胞相互作用和生物分子功能中起着核心作用。在自然界的许多例子中，病毒在受感染的细胞中自我组装，准备攻击其他细胞；指导胚胎形成特定器官和肢体的化学信号；以及通过化学信使交换在大脑中调解思想和记忆。在工业环境中，分子识别原理指导着新药、先进材料、催化剂和化学传感器的开发。然而，定量模型还没有达到设计针对特定伙伴的分子所需的原子分辨率和可靠性水平。分子根据形状和电荷等静态特性来识别其他分子。然而，它们也是动态实体，通过结合伙伴改变形状和采取互补构象的能力来进行识别。Gallicchio博士正在开发先进的技术和软件来针对这些复杂的特征，并捕捉分子运动和灵活性对分子识别过程的影响。这项研究是在化学本科生的积极参与下进行的，他们与合作的实验实验室合作，根据测量结果交互式地测试计算预测。软件和方法开发工作涉及物理、数学和计算机科学的学生。该项目的一个重要目标是帮助来自具有挑战性社会经济背景的学生成为下一代思维开阔的科学家，能够从多个融合的角度解决复杂问题。两种分子之间相互作用的可能性与标准自由结合能有关，或者，等价地，与双分子结合的平衡常数有关。该项目的目标是建立理论模型和计算算法，以准确有效地从第一性原理计算结合自由能，并了解自由能如何受到分子性质的影响。结合的统计分析理论将物理量（如结合位点的几何形状、配体的大小和它们相互作用的强度）与结合自由能模拟产生的动态数据联系起来。这一理论正被用于通过使用大规模平行非平衡协议来改进结合自由能方法，并基于统计模型估计的参数建立分子复合物的自动分类程序。为这个项目开发的代码作为高性能的、自动化的、免费的、文档完备的软件进行传播，支持各种平台和操作系统，包括计算网格和异构高性能计算系统。处理结合动力学方面的能力已应用于新化学合成，减轻化学危害，药物发现和材料设计。研究和教育活动在项目中紧密结合，为具有挑战性的社会经济背景的优秀学生提供科学和技术技能，以进入现代劳动力市场，同时为未来一代又一代的研究科学家播下种子。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Professor Emilio Gallicchio of Brooklyn College of the City University of New York is supported by an award from the Chemical Theory, Models, and Computational Methods and Chemistry of Life Processes programs in the Division of Chemistry to develop theoretical models, computational algorithms, and simulation software to study molecular recognition processes. Molecular recognition---the ability of molecules to recognize and bind specifically to other molecules---is a fundamental aspect of all physicochemical processes, and plays a central role in cellular interactions and biomolecular function. Among the many examples in nature are the self-assembly of viruses in infected cells preparing to attack other cells; chemical signals directing embryos to form specific organs and limbs; and the mediation of thought and memory in the brain through chemical messenger exchange. In industrial settings, the principles of molecular recognition guide the development of new drugs, advanced materials, catalysts and chemical sensors. However, quantitative models do not yet reach the level of atomic resolution and reliability necessary for designing molecules that target specific partners. Molecules recognize other molecules based on static properties such as shape and charge. However, they are also dynamical entities that perform recognition through the ability of binding partners to change shape and assume complementary conformations. Dr. Gallicchio is developing advanced techniques and software to target these complex features and capture the influence of molecular motion and flexibility on molecular recognition processes. The research is being carried out with active participation from Chemistry undergraduate students who, working with partnering experimental laboratories, interactively test computational predictions against measurements. The software and method development efforts involve Physics, Math, and Computer Science students. An important goal of the project is to help students from challenging socio-economic backgrounds to become the next generation of broad-thinking scientists, capable of tackling complex problems from multiple, convergent perspectives. The likelihood of interaction between two molecular species is related to the standard free energy of binding, or, equivalently, the equilibrium constant for bimolecular association. The goal of this project is to develop theoretical models and computational algorithms to accurately and efficiently compute binding free energies from first principles, and to understand how free energies are influenced by molecular properties. A statistical analytic theory of binding relates physical quantities such as the geometry of the binding site, the size of the ligand, and the strength of their interactions, to dynamical data produced from binding free energy simulations. This theory is being used to improve binding free energy methods through the use of massively-parallel non-equilibrium protocols, and to build an automated classification procedure for molecular complexes based on the parameters estimated from the statistical model. The code developed for this project is being disseminated as high-performance, automated, freely available, and well-documented software, supporting a variety of platforms and operating systems, including computational grids and heterogeneous high-performance computing systems. The ability to treat dynamical aspects of binding has application to novel chemical synthesis, mitigation of chemical hazards, drug discovery, and materials design. Research and educational activities are closely integrated within the project and are providing deserving students from challenging socio-economic backgrounds with the scientific and technological skills to enter the modern workforce, while planting the seeds for growing future generations of research scientists.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.