在蛋白-药物结合能计算中应用高精度量子化学方法如密度泛函理论可望显著提高计算精确度，但密度泛函理论的高标度计算量和巨大的蛋白药物复合体系构成了巨大挑战。该问题可由分块计算解决。然而现有文献报道主要关注分块后的“计算”部分，却鲜有涉及同样重要的“分块”方法——即将完整体系分为碎片所应遵循的规则。缺乏完善、普适、方便的蛋白药物复合体系分块方法导致蛋白药物结合能高精度量子化学分块计算目前只局限于个案研究，据规模化应用仍有较大差距。本项目拟在前期工作的基础上，基于药效团理论设计蛋白药物复合体系分块处理策略，探索能保证计算精度的分块规则，设计基于预测控制计算量的智能分块方式，编写自动化分块程序，来得到标准化、智能化和自动化的蛋白药物复合体系智能自动分块方法。研究成果可实现针对不同体系自动设计和应用最适合的分块方案得到普适、准确、实用的蛋白药物结合能计算方法，从而为药物研发提供高精度的理论计算辅助。

The accuracy of protein-drug binding energy calculation could be improved by the introduction of high level quantum chemistry theories such as density functional theory, but the dramatic computational cost of the combination of DFT and protein means a huge challenge. Such problem could be solved by the so called divide & conquer method. However, current literatures mainly focused on the ‘conquer’ part, while seldomly touch the also important ‘divide’ part, or how to fragment the protein drug complex. Lack of a convenient and universal protein drug complex fragment method limited the application of protein drug binding energy divide & conquer computation to case studies, far from wide range application. Based previous works, this project aims at develop a protein-drug complex fragment strategy according to pharmacophore theory. A smart, standardized and automatic protein drug complex fragment program will be obtained by exploring the rules which guarantee the calculation accuracy and setting up smart method for computational cost evaluation and control. The achievement of this project could lead to useful and practical protein drug binding energy computation methods which could provide advanced theoretical support for drug development.