Machine learning augmented molecular simulation pipelines for modelling allosteric modulation of protein function

机器学习增强分子模拟管道，用于模拟蛋白质功能的变构调节

• 批准号: 2871271
• 金额: --
• 依托单位: University of Edinburgh
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2871271
• 关键词: Machine; learning; augmented; molecular; simulation

This project will focus on developing computational methodologies grounded in molecular simulation and machine learning for predicting how the binding of a ligand to the surface of a protein will modulate its biological function. Such capability may facilitate the rational design of allosteric modulators of protein function which of enormous importance in drug discovery. This work builds on Markov State Modelling methodologies and alchemical free energy calculation methodologies that the Michel research group has recently developed to simulate large scale conformational changes in protein structures (Chem. Sci. , 11, 2670-2680, 2020) and to estimate protein-ligand binding affinities (Chem. Sci. , 13, 5220-5229, 2022). Throughout the project there will be opportunities to interact with pharmaceutical companies interested in such methodologies.

该项目将专注于开发基于分子模拟和机器学习的计算方法，用于预测配体与蛋白质表面的结合如何调节其生物学功能。这种能力可以促进蛋白质功能的变构调节剂的合理设计，这在药物发现中具有巨大的重要性。这项工作建立在马尔可夫状态建模方法和炼金术自由能计算方法的基础上，米歇尔研究小组最近开发了这些方法来模拟蛋白质结构中的大规模构象变化（Chem. Sci.，11，2670-2680，2020）和估计蛋白质-配体结合亲和力（Chem. Sci.，13，5220-5229，2022）。在整个项目过程中，将有机会与对这种方法感兴趣的制药公司进行互动。