• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利:CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验

机器学习增强的量子-经典结合自由能计算

Machine Learning-Enhanced Calculation of Quantum-Classical Binding Free Energies.

作者信息

Bensberg Moritz, Eckhoff Marco, Thomasen F Emil, Bro-Jørgensen William, Teynor Matthew S, Sora Valentina, Weymuth Thomas, Husistein Raphael T, Knudsen Frederik E, Krogh Anders, Lindorff-Larsen Kresten, Reiher Markus, Solomon Gemma C

机构信息

Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir-Prelog-Weg 2, Zurich 8093, Switzerland.

Department of Biology, Linderstrøm-Lang Centre for Protein Science, University of Copenhagen, Ole Maaløes Vej 5, Copenhagen N DK-2200, Denmark.

出版信息

J Chem Theory Comput. 2025 Aug 26;21(16):8182-8198. doi: 10.1021/acs.jctc.5c00388. Epub 2025 Aug 5.

DOI:10.1021/acs.jctc.5c00388
PMID:40762518
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC12392446/
Abstract

Binding free energies are key elements in understanding and predicting the strength of protein-drug interactions. While classical free energy simulations yield good results for many purely organic ligands, drugs, including transition metal atoms, often require quantum chemical methods for an accurate description. We propose a general and automated workflow that samples the potential energy surface with hybrid quantum mechanics/molecular mechanics (QM/MM) calculations and trains a machine learning (ML) potential on the QM/MM energies and forces to enable efficient alchemical free energy simulations. To represent systems including many different chemical elements efficiently and to account for the different descriptions of QM and MM atoms, we propose an extension of element-embracing atom-centered symmetry functions for QM/MM data as an ML descriptor. The ML potential approach takes electrostatic embedding and long-range electrostatics into account. We demonstrate the applicability of the workflow on the well-studied protein-ligand complex of myeloid cell leukemia 1 and the inhibitor 19G and on the anticancer drug NKP1339 acting on the glucose-regulated protein 78.

摘要

结合自由能是理解和预测蛋白质-药物相互作用强度的关键要素。虽然经典自由能模拟对于许多纯有机配体能产生良好结果,但包括过渡金属原子的药物通常需要量子化学方法才能进行准确描述。我们提出了一种通用且自动化的工作流程,该流程通过混合量子力学/分子力学(QM/MM)计算对势能面进行采样,并基于QM/MM能量和力训练机器学习(ML)势能,以实现高效的炼金术自由能模拟。为了有效地表示包含许多不同化学元素的系统,并考虑QM和MM原子的不同描述,我们提出将用于QM/MM数据的包含元素的原子中心对称函数扩展为一种ML描述符。ML势能方法考虑了静电嵌入和长程静电作用。我们在研究充分的髓系细胞白血病1与抑制剂19G的蛋白质-配体复合物以及作用于葡萄糖调节蛋白78的抗癌药物NKP1339上证明了该工作流程的适用性。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/e884d55968db/ct5c00388_0007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/998bbc1533ed/ct5c00388_0001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/cad4c3cc5fbd/ct5c00388_0002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/e39b4cfed7a3/ct5c00388_0003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/8a822182e464/ct5c00388_0004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/69ee382385c0/ct5c00388_0005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/573d16b2d87d/ct5c00388_0006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/e884d55968db/ct5c00388_0007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/998bbc1533ed/ct5c00388_0001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/cad4c3cc5fbd/ct5c00388_0002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/e39b4cfed7a3/ct5c00388_0003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/8a822182e464/ct5c00388_0004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/69ee382385c0/ct5c00388_0005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/573d16b2d87d/ct5c00388_0006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/813d/12392446/e884d55968db/ct5c00388_0007.jpg

相似文献

1
Machine Learning-Enhanced Calculation of Quantum-Classical Binding Free Energies.机器学习增强的量子-经典结合自由能计算
J Chem Theory Comput. 2025 Aug 26;21(16):8182-8198. doi: 10.1021/acs.jctc.5c00388. Epub 2025 Aug 5.
2
Efficient Free Energies from a Simplified Electrostatic Embedding QM/MM Approach Based on Electrostatic Potential Fitted Operators.基于静电势拟合算符的简化静电嵌入量子力学/分子力学方法得到的有效自由能
J Chem Inf Model. 2025 Sep 8;65(17):9196-9207. doi: 10.1021/acs.jcim.5c01563. Epub 2025 Aug 26.
3
Reproducibility of QM/MM Calculations for the SARS-CoV-2 Main Protease.严重急性呼吸综合征冠状病毒2型主要蛋白酶的量子力学/分子力学计算的可重复性
J Chem Theory Comput. 2025 Jul 24. doi: 10.1021/acs.jctc.5c00841.
4
Liquid Phase Modeling in Porous Media: Adsorption of Methanol and Ethanol in H-MFI in Condensed Water.多孔介质中的液相建模:甲醇和乙醇在冷凝水中于H-MFI中的吸附
J Chem Theory Comput. 2025 Jun 24;21(12):6121-6134. doi: 10.1021/acs.jctc.5c00427. Epub 2025 Jun 11.
5
QM/CG-MM: Systematic Embedding of Quantum Mechanical Systems in a Coarse-Grained Environment with Accurate Electrostatics.QM/CG-MM:在具有精确静电作用的粗粒度环境中对量子力学系统进行系统嵌入。
J Phys Chem A. 2024 Jul 25;128(29):6061-6071. doi: 10.1021/acs.jpca.4c02906. Epub 2024 Jul 17.
6
Solvation Free Energies of Drug-like Molecules via Fast Growth in an Explicit Solvent: Assessment of the AM1-BCC, RESP/HF/6-31G*, RESP-QM/MM, and ABCG2 Fixed-Charge Approaches.通过在显式溶剂中的快速增长计算类药物分子的溶剂化自由能:对AM1-BCC、RESP/HF/6-31G*、RESP-QM/MM和ABCG2固定电荷方法的评估
J Chem Theory Comput. 2025 Aug 26;21(16):7977-7990. doi: 10.1021/acs.jctc.5c00749. Epub 2025 Aug 11.
7
Basic Stability Tests of Machine Learning Potentials for Molecular Simulations in Computational Drug Discovery.计算药物发现中分子模拟机器学习势的基本稳定性测试
J Chem Inf Model. 2025 Sep 8;65(17):8980-8999. doi: 10.1021/acs.jcim.5c01150. Epub 2025 Aug 27.
8
Impact of Fluctuations in the Peridinin-Chlorophyll a-Protein on the Energy Transfer: Insights from Classical and QM/MM Molecular Dynamics Simulations.叶绿素a-蛋白波动对能量转移的影响:来自经典和QM/MM分子动力学模拟的见解
Biochemistry. 2025 Feb 18;64(4):879-894. doi: 10.1021/acs.biochem.4c00568. Epub 2025 Feb 4.
9
VMD as a Platform for Interactive Small Molecule Preparation and Visualization in Quantum and Classical Simulations.VMD 作为一个平台,用于在量子和经典模拟中进行交互式小分子准备和可视化。
J Chem Inf Model. 2023 Aug 14;63(15):4664-4678. doi: 10.1021/acs.jcim.3c00658. Epub 2023 Jul 28.
10
SparcleQC: Automated Input File Creation for QM/MM Studies of Protein:Ligand Complexes.SparcleQC:用于蛋白质-配体复合物QM/MM研究的自动输入文件创建
J Chem Inf Model. 2025 Jul 14;65(13):6433-6440. doi: 10.1021/acs.jcim.5c00617. Epub 2025 Jun 17.

引用本文的文献

1
Hierarchical Quantum Embedding by Machine Learning for Large Molecular Assemblies.通过机器学习实现的用于大分子组装体的分层量子嵌入
J Chem Theory Comput. 2025 Aug 12;21(15):7662-7674. doi: 10.1021/acs.jctc.5c00389. Epub 2025 Jul 29.

本文引用的文献

1
Hierarchical Quantum Embedding by Machine Learning for Large Molecular Assemblies.通过机器学习实现的用于大分子组装体的分层量子嵌入
J Chem Theory Comput. 2025 Aug 12;21(15):7662-7674. doi: 10.1021/acs.jctc.5c00389. Epub 2025 Jul 29.
2
NepoIP/MM: Toward Accurate Biomolecular Simulation with a Machine Learning/Molecular Mechanics Model Incorporating Polarization Effects.NepoIP/MM:借助包含极化效应的机器学习/分子力学模型实现精确的生物分子模拟。
J Chem Theory Comput. 2025 Jun 10;21(11):5588-5598. doi: 10.1021/acs.jctc.5c00372. Epub 2025 May 21.
3
DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials.
深度势能工具包v3:一种用于机器学习势能的多后端框架。
J Chem Theory Comput. 2025 May 13;21(9):4375-4385. doi: 10.1021/acs.jctc.5c00340. Epub 2025 May 2.
4
Multiscale Embedding for Quantum Computing.用于量子计算的多尺度嵌入
J Chem Theory Comput. 2025 May 13;21(9):4591-4603. doi: 10.1021/acs.jctc.5c00241. Epub 2025 Apr 15.
5
Neural Network Potential with Multiresolution Approach Enables Accurate Prediction of Reaction Free Energies in Solution.采用多分辨率方法的神经网络势能够精确预测溶液中的反应自由能。
J Am Chem Soc. 2025 Feb 26;147(8):6835-6856. doi: 10.1021/jacs.4c17015. Epub 2025 Feb 17.
6
Evaluation of Machine Learning/Molecular Mechanics End-State Corrections with Mechanical Embedding to Calculate Relative Protein-Ligand Binding Free Energies.利用机械嵌入评估机器学习/分子力学终态校正以计算相对蛋白质-配体结合自由能
J Chem Theory Comput. 2025 Jan 28;21(2):967-977. doi: 10.1021/acs.jctc.4c01427. Epub 2025 Jan 3.
7
Software Infrastructure for Next-Generation QM/MM-ΔMLP Force Fields.用于下一代QM/MM-ΔMLP力场的软件基础设施。
J Phys Chem B. 2024 Jul 4;128(26):6257-6271. doi: 10.1021/acs.jpcb.4c01466. Epub 2024 Jun 21.
8
Current State of Open Source Force Fields in Protein-Ligand Binding Affinity Predictions.开源力场在蛋白质-配体结合亲和力预测中的现状。
J Chem Inf Model. 2024 Jul 8;64(13):5063-5076. doi: 10.1021/acs.jcim.4c00417. Epub 2024 Jun 19.
9
SCINE-Software for chemical interaction networks.SCINE - 化学相互作用网络软件。
J Chem Phys. 2024 Jun 14;160(22). doi: 10.1063/5.0206974.
10
Learning QM/MM potential using equivariant multiscale model.使用等变多尺度模型学习量子力学/分子力学势能
J Chem Phys. 2024 Jun 7;160(21). doi: 10.1063/5.0205123.