New Algorithms for Large Time-Step Molecular Dynamics Simulations and their Application to Protein and Nucleic Acids

大时间步长分子动力学模拟的新算法及其在蛋白质和核酸中的应用

• 批准号: 9310295
• 负责人: Tamar Schlick
• 金额: $ 4.62万
• 依托单位: New York University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1993
• 资助国家: 美国
• 起止时间: 1993-09-01 至 1996-02-29
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9310295&HistoricalAwards=false
• 关键词: New; Algorithms; Large; Time; Step

9310295 Schlick The goal in this proposal is to extend this work further so that is can be applied routinely to protein and nucleic acids. We propose to develop and implement a new algorithm that combines implicit-integration and normal-mode techniques. The new algorithm will maintain the large time-step stability inherited in the implicit scheme, but also resolve the high-frequency components of the motion so that the activities of high-frequency modes can be retained. With parallel implementations on state-of-the-art machines, this will make possible simulations at time-steps of order 100 femtoseconds; the total simulation time will then approach the millisecond range.

9310295 石里克 本提案的目标是进一步扩展这项工作，使其可以常规地应用于蛋白质和核酸。 我们建议开发和实现一个新的算法，结合隐式积分和正常模式技术。 新算法既保持了隐式格式的大时间步长稳定性，又能分解运动的高频分量，从而保留了高频模态的活动性。 通过在最先进的机器上并行实现，这将使100飞秒量级的时间步长的模拟成为可能;总的模拟时间将接近毫秒范围。