• 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

预测化学：用于反应部署、反应开发和反应发现的机器学习

Predictive chemistry: machine learning for reaction deployment, reaction development, and reaction discovery.

作者信息

Tu Zhengkai, Stuyver Thijs, Coley Connor W

机构信息

Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge MA 02139 USA

Department of Chemical Engineering, Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge MA 02139 USA.

出版信息

Chem Sci. 2022 Nov 28;14(2):226-244. doi: 10.1039/d2sc05089g. eCollection 2023 Jan 4.

DOI:10.1039/d2sc05089g
PMID:36743887
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC9811563/
Abstract

The field of predictive chemistry relates to the development of models able to describe how molecules interact and react. It encompasses the long-standing task of computer-aided retrosynthesis, but is far more reaching and ambitious in its goals. In this review, we summarize several areas where predictive chemistry models hold the potential to accelerate the deployment, development, and discovery of organic reactions and advance synthetic chemistry.

摘要

预测化学领域涉及能够描述分子如何相互作用和反应的模型的开发。它涵盖了计算机辅助逆合成这一长期任务，但其目标更为广泛和宏大。在本综述中，我们总结了预测化学模型在加速有机反应的应用、开发和发现以及推动合成化学发展方面具有潜力的几个领域。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/551de5831584/d2sc05089g-p3.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/5328b7926ecd/d2sc05089g-f1.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/9480e7917615/d2sc05089g-f2.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/f54c4297500f/d2sc05089g-f3.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/7c1741beb5da/d2sc05089g-f4.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/9a9c129e4a0f/d2sc05089g-f5.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/e6c14ee79403/d2sc05089g-f6.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/e9abd0666fcb/d2sc05089g-f7.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/f07dc58fc82e/d2sc05089g-p1.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/f6bdc3f87fd4/d2sc05089g-p2.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/551de5831584/d2sc05089g-p3.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/5328b7926ecd/d2sc05089g-f1.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/9480e7917615/d2sc05089g-f2.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/f54c4297500f/d2sc05089g-f3.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/7c1741beb5da/d2sc05089g-f4.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/9a9c129e4a0f/d2sc05089g-f5.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/e6c14ee79403/d2sc05089g-f6.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/e9abd0666fcb/d2sc05089g-f7.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/f07dc58fc82e/d2sc05089g-p1.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/f6bdc3f87fd4/d2sc05089g-p2.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3101/9811563/551de5831584/d2sc05089g-p3.jpg

相似文献

1
Predictive chemistry: machine learning for reaction deployment, reaction development, and reaction discovery.预测化学：用于反应部署、反应开发和反应发现的机器学习
Chem Sci. 2022 Nov 28;14(2):226-244. doi: 10.1039/d2sc05089g. eCollection 2023 Jan 4.
2
Chemistry-informed molecular graph as reaction descriptor for machine-learned retrosynthesis planning.基于化学信息的分子图作为机器学习逆向合成规划的反应描述符。
Proc Natl Acad Sci U S A. 2022 Oct 11;119(41):e2212711119. doi: 10.1073/pnas.2212711119. Epub 2022 Oct 3.
3
RetroComposer: Composing Templates for Template-Based Retrosynthesis Prediction.RetroComposer：基于模板的反合成预测的模板作曲。
Biomolecules. 2022 Sep 19;12(9):1325. doi: 10.3390/biom12091325.
4
Data-driven Chemical Reaction Prediction and Retrosynthesis.数据驱动的化学反应预测与逆合成
Chimia (Aarau). 2019 Dec 18;73(12):997-1000. doi: 10.2533/chimia.2019.997.
5
Machine-Learning-Guided Discovery of Electrochemical Reactions.机器学习指导下的电化学反应发现。
J Am Chem Soc. 2022 Dec 14;144(49):22599-22610. doi: 10.1021/jacs.2c08997. Epub 2022 Dec 2.
6
RPBP: Deep Retrosynthesis Reaction Prediction Based on Byproducts.RPBP：基于副产物的深度逆合成反应预测
J Chem Inf Model. 2023 Oct 9;63(19):5956-5970. doi: 10.1021/acs.jcim.3c00274. Epub 2023 Sep 19.
7
Machine Learning in Computer-Aided Synthesis Planning.计算机辅助合成规划中的机器学习
Acc Chem Res. 2018 May 15;51(5):1281-1289. doi: 10.1021/acs.accounts.8b00087. Epub 2018 May 1.
8
Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction.用于逆合成和反应预测的神经符号机器学习
Chemistry. 2017 May 2;23(25):5966-5971. doi: 10.1002/chem.201605499. Epub 2017 Feb 22.
9
Fuelling the Digital Chemistry Revolution with Language Models.用语言模型推动数字化学革命。
Chimia (Aarau). 2023 Aug 9;77(7-8):484-488. doi: 10.2533/chimia.2023.484.
10
Controlling an organic synthesis robot with machine learning to search for new reactivity.用机器学习控制有机合成机器人来寻找新的反应性。
Nature. 2018 Jul;559(7714):377-381. doi: 10.1038/s41586-018-0307-8. Epub 2018 Jul 18.

引用本文的文献

1
Electron flow matching for generative reaction mechanism prediction.用于生成反应机理预测的电子流匹配
Nature. 2025 Aug 20. doi: 10.1038/s41586-025-09426-9.
2
Discovery of Ni Complexes for CO Insertion Enabled by a Machine Learning-Computational-Selection Sequence.通过机器学习-计算-筛选序列实现的用于CO插入的镍配合物的发现
J Am Chem Soc. 2025 Jul 30;147(30):26149-26157. doi: 10.1021/jacs.5c00441. Epub 2025 Jul 18.
3
Unveiling Unusual Reactivity of SO and Unusual Type of S-X Long Bonds.揭示SO的异常反应性及S-X长键的异常类型。

本文引用的文献

1
Simultaneously improving reaction coverage and computational cost in automated reaction prediction tasks.在自动反应预测任务中同时提高反应覆盖率和计算成本。
Nat Comput Sci. 2021 Jul;1(7):479-490. doi: 10.1038/s43588-021-00101-3. Epub 2021 Jul 22.
2
Retrosynthetic planning with experience-guided Monte Carlo tree search.基于经验引导的蒙特卡洛树搜索的逆合成规划
Commun Chem. 2023 Jun 10;6(1):120. doi: 10.1038/s42004-023-00911-8.
3
GRetro as a two-step graph generative models for retrosynthesis prediction.GRetro作为一种用于逆合成预测的两步图生成模型。
Inorg Chem. 2025 Jul 21;64(28):14684-14692. doi: 10.1021/acs.inorgchem.5c02435. Epub 2025 Jul 8.
4
Developing a Machine Learning Model for Hydrogen Bond Acceptance Based on Natural Bond Orbital Descriptors.基于自然键轨道描述符开发用于氢键接受的机器学习模型。
J Org Chem. 2025 Jul 18;90(28):9776-9788. doi: 10.1021/acs.joc.5c00724. Epub 2025 Jul 6.
5
Benchmarking molecular conformer augmentation with context-enriched training: graph-based transformer versus GNN models.基于上下文丰富训练的分子构象增强基准测试：基于图的变压器模型与图神经网络模型对比
J Cheminform. 2025 May 22;17(1):80. doi: 10.1186/s13321-025-01004-5.
6
ASKCOS: Open-Source, Data-Driven Synthesis Planning.ASKCOS：开源、数据驱动的合成规划。
Acc Chem Res. 2025 Jun 3;58(11):1764-1775. doi: 10.1021/acs.accounts.5c00155. Epub 2025 May 21.
7
Data accessibility in the chemical sciences: an analysis of recent practice in organic chemistry journals.化学科学中的数据可获取性：对有机化学期刊近期实践的分析
Beilstein J Org Chem. 2025 May 2;21:864-876. doi: 10.3762/bjoc.21.70. eCollection 2025.
8
Challenging Reaction Prediction Models to Generalize to Novel Chemistry.挑战反应预测模型以推广至新化学领域。
ACS Cent Sci. 2025 Mar 12;11(4):539-549. doi: 10.1021/acscentsci.5c00055. eCollection 2025 Apr 23.
9
A meta-learning approach for selectivity prediction in asymmetric catalysis.一种用于不对称催化中选择性预测的元学习方法。
Nat Commun. 2025 Apr 16;16(1):3599. doi: 10.1038/s41467-025-58854-8.
10
Unified Deep Learning of Molecular and Protein Language Representations with T5ProtChem.利用T5ProtChem对分子和蛋白质语言表征进行统一深度学习。
J Chem Inf Model. 2025 Apr 28;65(8):3990-3998. doi: 10.1021/acs.jcim.5c00051. Epub 2025 Apr 8.
Commun Chem. 2023 May 30;6(1):102. doi: 10.1038/s42004-023-00897-3.
4
OSCAR: an extensive repository of chemically and functionally diverse organocatalysts.OSCAR：一个包含化学和功能多样的有机催化剂的广泛储存库。
Chem Sci. 2022 Oct 25;13(46):13782-13794. doi: 10.1039/d2sc04251g. eCollection 2022 Nov 30.
5
A broadly applicable quantitative relative reactivity model for nucleophilic aromatic substitution (SAr) using simple descriptors.一种使用简单描述符的适用于亲核芳香取代（SAr）的广泛适用的定量相对反应性模型。
Chem Sci. 2022 Oct 17;13(43):12681-12695. doi: 10.1039/d2sc04041g. eCollection 2022 Nov 9.
6
Machine learning meets mechanistic modelling for accurate prediction of experimental activation energies.机器学习与机理建模相结合以精确预测实验活化能。
Chem Sci. 2020 Nov 5;12(3):1163-1175. doi: 10.1039/d0sc04896h. eCollection 2021 Jan 21.
7
Root-aligned SMILES: a tight representation for chemical reaction prediction.根对齐的简化分子线性输入规范（SMILES）：一种用于化学反应预测的紧凑表示法。
Chem Sci. 2022 Jul 12;13(31):9023-9034. doi: 10.1039/d2sc02763a. eCollection 2022 Aug 10.
8
Machine Learning Yield Prediction from NiCOlit, a Small-Size Literature Data Set of Nickel Catalyzed C-O Couplings.机器学习从 NiCOlit 中预测产率，NiCOlit 是一个镍催化 C-O 偶联的小规模文献数据集。
J Am Chem Soc. 2022 Aug 17;144(32):14722-14730. doi: 10.1021/jacs.2c05302. Epub 2022 Aug 8.
9
Permutation Invariant Graph-to-Sequence Model for Template-Free Retrosynthesis and Reaction Prediction.无模板回溯合成和反应预测的置换不变图到序列模型。
J Chem Inf Model. 2022 Aug 8;62(15):3503-3513. doi: 10.1021/acs.jcim.2c00321. Epub 2022 Jul 26.
10
Using Active Learning to Develop Machine Learning Models for Reaction Yield Prediction.利用主动学习开发用于反应产率预测的机器学习模型。
Mol Inform. 2022 Dec;41(12):e2200043. doi: 10.1002/minf.202200043. Epub 2022 Jul 14.