膜蛋白作为细胞的核心组成部分担负细胞内外物质交换等重要功能,是药物研发的最大一类靶点。目前对于复杂膜蛋白的结构解析、物质输运功能的微观机理以及金属离子对其功能影响的高精度理解和认识都非常困难和有限，主要因为目前国际上还没有针对细胞膜重要组分：磷脂和胆固醇的高精度全原子可极化分子力场。同时，现有全原子可极化分子力场进行分子动力学模拟时，其计算速度要比传统力场慢至少1个数量级；而且没有与高效增强型采样技术相融合。所以，无法用于复杂生物体系的高精度理论研究。本项目立足于已有研究基础，首次提出建立所有种类细胞膜磷脂和胆固醇的全原子可极化AMOEBA分子力场，结合GPU加速和增强采样技术Metadynamics，实现高精度分子模拟新方法，并针对重要膜蛋白的结构解析、行使功能的微观机理以及金属离子对其功能的调控机制进行系统理论研究，为复杂生物体系的高精度模拟以及药物研发提供强有力的理论支撑和研究工具。

As an indispensable member in the cell, membrane proteins, in addition to make biological functions by transporting species across the cell membrane, are the largest type of targets for drug discovery. The structural determination, accurate understanding of mechanism of transportation of membrane proteins and the effects of metal ions on the functions of membrane proteins are very hard to be studied and limited. The main reason for this is because there is no all-atom polarizable force field (PFF) available for accurate studying the key components of cell membrane: lipids and cholesterols. Meanwhile, running speed of molecular dynamic with all-atom PFF is at least one-order slower than traditional force field. Also, as now there is no efficient enhanced-sampling techniques available in the program using PFF. Thus, the accurate MD studies of complicated biomolecules cannot be done using PFF till now. Based on our research experience and preliminary results, we propose to construct all-atom PFF (AMOEBA) for lipid molecules and cholesterols, combine with GPU-based hardware acceleration and metadynamics-based enhanced sampling technique, and finally deliver a new and effective MD simulation method with PFF for complex systems. With the new developed methodology we will help the structural determination of membrane proteins, study systematically the microscopic mechanism of membrane protein transportation and the effects of metal ions on the functions of membrane proteins. This will provide solid theoretical basis and new research tool for understanding the functional mechanism of complicate biomolecules, and drug discovery.