机器学习 量子力学/分子力学潜能:评估二叶酸减少酶催化反应中的可转移性
Abdul Raafik Arattu Thodika1, Xiaoliang Pan2, Yihan Shao2
1Department of Chemistry and Biochemistry, University of Texas at Arlington, Arlington, Texas 76019, United States.
Journal of chemical theory and computation
|January 15, 2025
概括
机器学习潜力 (MLP) 可以在不需要再培训的情况下预测酶催化. 一个预训练的ΔMLP模型显示在酶突变中具有良好的可转移性,但当移动到水环境时存在限制.
科学领域:
- 计算化学是一种计算化学.
- 生物化学 生物化学
- 机器学习是机器学习.
背景情况:
- 将机器学习潜力 (MLP) 与量子力学/分子力学 (QM/MM) 模拟集成,为研究酶催化提供了一种强大的方法.
- 通过QM/MM模拟生成MLP的训练数据是耗时且系统特定的,阻碍了实际应用.
研究的目的:
- 评估预训练的ΔMLP模型在不同酶突变和环境中的可转移性.
- 评估是否需要为新的酶系统或变体重新培训MLP.
主要方法:
- 使用基于QM/MM的ML架构来测试预训练的ΔMLP模型.
- 在单点替代,同源酶和水性环境中评估了可转移性.
- 与MLP再培训和没有MLP再培训的比较的自由能量概况.
主要成果:
- 在没有重新训练的情况下,ΔMLP模型准确地预测了酶突变对静电相互作用和自由能量概况的影响.
- 在各种酶突变和同源酶中,可转移性很强.
- 在过渡到富含水的分子力学环境时,观察到可转移性的显著限制.
结论:
- 基于QM/MM的ML架构证明了对各种酶系统的稳定性.
- 预先训练有素的MLP可以减少研究酶催化物的计算负担.
- 需要进一步的研究来改善MLP的可转移性,特别是对于酶到溶剂的转换.
更多相关视频
05:51Isotopic Effect in Double Proton Transfer Process of Porphycene Investigated by Enhanced QM/MM Method
Published on: July 19, 2019
6.2K
10:29Quantitative Structure-Activity Relationship, Activity Prediction, and Molecular Dynamics of Non-nucleotide Reverse Transcriptase Inhibitors
Published on: May 9, 2025
344
相关概念视频
03:25Free Energy Changes for Nonstandard States
The free energy change for a process taking place with reactants and products present under nonstandard conditions (pressures other than 1 bar; concentrations other than 1 M) is related to the standard free energy change according to this equation:
01:30Quantitative Aspects of Drug-Receptor Interaction
The receptor occupancy theory connects a drug's response to the number of occupied receptors. With higher drug concentrations, more receptors are occupied, leading to increased responses. The formation of drug-receptor complexes involves association and dissociation rates, which reach equilibrium when the forward and backward reactions are equal. The equilibrium association constant (Ka) and its inverse, the equilibrium dissociation constant (Kd), indicate drug affinity. Higher Ka and lower Kd...