Using Machine Learning To Predict Suitable Conditions for Organic Reactions
Open accessAbstract
Reaction condition recommendation is an essential element for the realization of computer-assisted synthetic planning. Accurate suggestions of reaction conditions are required for experimental validation and can have a significant effect on the success or failure of an attempted transformation. However, de novo condition recommendation remains a challenging and under-explored problem and relies heavily on chemists' knowledge and experience. In this work, we develop a neural-network model to predict the chemical context (catalyst(s), solvent(s), reagent(s)), as well as the temperature most suitable for any particular organic reaction. Trained on ∼10 million examples from Reaxys, the model is able to propose conditions where a close match to the recorded catalyst, solvent, and reagent is found within the top-10 predictions 69.6% of the time, with top-10 accuracies for individual species reaching 80-90%. Temperature is accurately predicted within ±20 °C from the recorded temperature in 60-70% of test cases, with higher accuracy for cases with correct chemical context predictions. The utility of the model is illustrated through several examples spanning a range of common reaction classes. We also demonstrate that the model implicitly learns a continuous numerical embedding of solvent and reagent species that captures their functional similarity.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Gao et al. reported NN-based models trained on ~10 million reactions from Reaxys that identify appropriate catalysts, solvents, reagents, and temperatures for specified reactions.
“Gao et al.163 reported NN-based models trained on ~10 million reactions from Reaxys that identify appropriate catalysts, solvents, reagents, and temperatures for a specified reactions.”
From QSAR without borders (2020) · cites this paper as [163]
A 70% match with experimental conditions was found within the top-10 predictions of Gao et al.'s models.
“A 70% match with experimental conditions was found within the top-10 predictions.”
From QSAR without borders (2020) · cites this paper as [163]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Distributed Representations of Words and Phrases and their Compositionality
Mikolov, T., Sutskever, I., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1310.4546 · Cited by 24,533 · see citing works
A Learning Algorithm for Continually Running Fully Recurrent Neural Networks
Williams, R. J., & Zipser, D. (1989) · Neural Computation · doi:10.1162/neco.1989.1.2.270 · Cited by 4,369 · see citing works
Kamlet, M. J., Abboud, J. ‐. M., Abraham, M. H., et al. (1983) · The Journal of Organic Chemistry · doi:10.1021/jo00165a018 · Cited by 3,624 · see citing works
Planning chemical syntheses with deep neural networks and symbolic AI
Segler, M. H. S., Preuss, M., & Waller, M. P. (2018) · Nature · doi:10.1038/nature25978 · Cited by 1,847 · see citing works
GARCÍA-ALONSO, C. R., Pérez, L. M., & Fernández, J. C. F. (2011) · Annals of Operations Research · doi:10.1007/s10479-011-0841-3 · Cited by 1,357 · see citing works
Prediction of Organic Reaction Outcomes Using Machine Learning
Coley, C. W., Barzilay, R., Jaakkola, T., Green, W. H., & Jensen, K. F. (2017) · ACS Central Science · doi:10.1021/acscentsci.7b00064 · Cited by 755 · see citing works
Self-citationMachine Learning in Computer-Aided Synthesis Planning
Coley, C. W., Green, W. H., & Jensen, K. F. (2018) · Accounts of Chemical Research · doi:10.1021/acs.accounts.8b00087 · Cited by 704 · see citing works
Self-citationComputer-Assisted Design of Complex Organic Syntheses
Corey, E. J., & Wipke, W. T. (1969) · Science · doi:10.1126/science.166.3902.178 · Cited by 688 · see citing works
Computer‐Assisted Synthetic Planning: The End of the Beginning
Szymkuć, S., Gajewska, E. P., Klucznik, T., Molga, K., Dittwald, P., Startek, M., Bajczyk, M., & Grzybowski, B. A. (2016) · Angewandte Chemie International Edition · doi:10.1002/anie.201506101 · Cited by 611 · see citing works
Neural‐Symbolic Machine Learning for Retrosynthesis and Reaction Prediction
Segler, M. H. S., & Waller, M. P. (2017) · Chemistry – A European Journal · doi:10.1002/chem.201605499 · Cited by 548 · see citing works
Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models
Liu, B., Ramsundar, B., Kawthekar, P., Shi, J., Gomes, J., Luu Nguyen, Q., Ho, S., Sloane, J., Wender, P., & Pande, V. (2017) · ACS Central Science · doi:10.1021/acscentsci.7b00303 · Cited by 536 · see citing works
LXIII.—A synthesis of tropinone
Robinson, R. C. (1917) · Journal of the Chemical Society Transactions · doi:10.1039/ct9171100762 · Cited by 532 · see citing works
Schwaller, P., Gaudin, T., Lányi, D., Bekas, C., & Laino, T. (2018) · Chemical Science · doi:10.1039/c8sc02339e · Cited by 449 · see citing works
Computer-Assisted Retrosynthesis Based on Molecular Similarity
Coley, C. W., Rogers, L., Green, W. H., & Jensen, K. F. (2017) · ACS Central Science · doi:10.1021/acscentsci.7b00355 · Cited by 403 · see citing works
Self-citationFeedback in Flow for Accelerated Reaction Development
Reizman, B. J., & Jensen, K. F. (2016) · Accounts of Chemical Research · doi:10.1021/acs.accounts.6b00261 · Cited by 261 · see citing works
Self-citationA self optimizing synthetic organic reactor system using real-time in-line NMR spectroscopy
Sans, V., Porwol, L., Dragone, V., & Cronin, L. (2014) · Chemical Science · doi:10.1039/c4sc03075c · Cited by 252 · see citing works
Route Designer: A Retrosynthetic Analysis Tool Utilizing Automated Retrosynthetic Rule Generation
Law, J. B., Zsoldos, Z., Simon, A., Reid, D., Liu, Y., Khew, S. Y., Johnson, A. P., Major, S., Wade, R. A., & Ando, H. Y. (2009) · Journal of Chemical Information and Modeling · doi:10.1021/ci800228y · Cited by 228 · see citing works
Schneider, N., Lowe, D. M., Sayle, R. A., & Landrum, G. A. (2014) · Journal of Chemical Information and Modeling · doi:10.1021/ci5006614 · Cited by 196 · see citing works
Modelling Chemical Reasoning to Predict and Invent Reactions
Segler, M. H. S., & Waller, M. P. (2016) · Chemistry - A European Journal · doi:10.1002/chem.201604556 · Cited by 196 · see citing works
Learning to Predict Chemical Reactions
Kayala, M. A., Azencott, C., Chen, J. H., & Baldi, P. (2011) · Journal of Chemical Information and Modeling · doi:10.1021/ci200207y · Cited by 190 · see citing works
Kayala, M. A., & Baldi, P. (2012) · Journal of Chemical Information and Modeling · doi:10.1021/ci3003039 · Cited by 186 · see citing works
Computational prediction of chemical reactions: current status and outlook
Engkvist, O., Norrby, P., Selmi, N., Lam, Y., Peng, Z., Sherer, E. C., Amberg, W., Erhard, T., & Smyth, L. A. (2018) · Drug Discovery Today · doi:10.1016/j.drudis.2018.02.014 · Cited by 185 · see citing works
Computer-aided molecular design of solvents for accelerated reaction kinetics
Struebing, H., Ganase, Z., Karamertzanis, P. G., et al. (2013) · Nature Chemistry · doi:10.1038/nchem.1755 · Cited by 181 · see citing works
Predicting Organic Reaction Outcomes with Weisfeiler-Lehman NetworkPreprint
Jin, W., Coley, C. W., Barzilay, R., & Jaakkola, T. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1709.04555 · Cited by 176 · see citing works
Self-citationWarr, W. A. (2014) · Molecular Informatics · doi:10.1002/minf.201400052 · Cited by 143 · see citing works
Bøgevig, A., Federsel, H., Huerta, F., Hutchings, M. G., Kraut, H., Langer, T., Löw, P., Oppawsky, C., Rein, T., & Saller, H. (2015) · Organic Process Research & Development · doi:10.1021/op500373e · Cited by 137 · see citing works
Online quantitative mass spectrometry for the rapid adaptive optimisation of automated flow reactors
Holmes, N., Akien, G. R., Savage, R. J. D., Stanetty, C., Baxendale, I. R., Blacker, A. J., Taylor, B. A., Woodward, R. L., Meadows, R. E., & Bourne, R. A. (2015) · Reaction Chemistry & Engineering · doi:10.1039/c5re00083a · Cited by 127 · see citing works
Computer‐aided synthesis design: 40 years on
Cook, A. P. F., Johnson, A. P., Law, J. B., Mirzazadeh, M., Ravitz, O., & Simon, A. (2011) · Wiley Interdisciplinary Reviews Computational Molecular Science · doi:10.1002/wcms.61 · Cited by 115 · see citing works
Simultaneous solvent screening and reaction optimization in microliter slugs
Reizman, B. J., & Jensen, K. F. (2015) · Chemical Communications · doi:10.1039/c5cc03651h · Cited by 97 · see citing works
Self-citationWang, H., & Yu, S. (2015) · Organic Letters · doi:10.1021/acs.orglett.5b01960 · Cited by 94 · see citing works
Baumgartner, L. M., Coley, C. W., Reizman, B. J., Gao, K. W., & Jensen, K. F. (2018) · Reaction Chemistry & Engineering · doi:10.1039/c8re00032h · Cited by 94 · see citing works
Self-citationFacile Assembly of Fused Benzo[4,5]furo Heterocycles
Liu, J. L., Fitzgerald, A. E., & Mani, N. S. (2008) · The Journal of Organic Chemistry · doi:10.1021/jo8000595 · Cited by 91 · see citing works
Expert System for Predicting Reaction Conditions: The Michael Reaction Case
Marcou, G., Aires de Sousa, J., Latino, D. A. R. S., de Luca, A., Horvath, D., Rietsch, V., & Varnek, A. (2015) · Journal of Chemical Information and Modeling · doi:10.1021/ci500698a · Cited by 86 · see citing works
Genetic optimization of combinatorial libraries
Gobbi, A., & Poppinger, D. (1998) · Biotechnology and Bioengineering · doi:10.1002/(sici)1097-0290(199824)61:1 · Cited by 77 · see citing works
Schaub, C., Müller, B. H., & Schmidt, R. R. (1998) · Glycoconjugate Journal · doi:10.1023/a:1006917717161 · Cited by 63 · see citing works
Sequential Pd(II)−Pd(0) Catalysis for the Rapid Synthesis of Coumarins
Li, K., Zeng, Y., Neuenswander, B., et al. (2005) · The Journal of Organic Chemistry · doi:10.1021/jo050671l · Cited by 61 · see citing works
Automatized Assessment of Protective Group Reactivity: A Step Toward Big Reaction Data Analysis
Lin, A. I., Madzhidov, T. I., Klimchuk, O., Nugmanov, R. I., Antipin, I. S., & Varnek, A. (2016) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.6b00319 · Cited by 51 · see citing works
Lautens, M., & Maddess, M. L. (2004) · Organic Letters · doi:10.1021/ol049883f · Cited by 49 · see citing works
Regioselective Arene Functionalization: Simple Substitution of Carboxylate by Alkyl Groups
Krüger, T., Vorndran, K., & Linker, T. (2009) · Chemistry - A European Journal · doi:10.1002/chem.200901774 · Cited by 39 · see citing works
Sialyltransferase Inhibitors Based on CMP-Quinic Acid
Schaub, C., Müller, B. H., & Schmidt, R. R. (2000) · European Journal of Organic Chemistry · doi:10.1002/(sici)1099-0690(200005)2000:9 · Cited by 30 · see citing works
Faroux-Corlay, B., Clary, L., Gadras, C., et al. (2000) · Carbohydrate Research · doi:10.1016/s0008-6215(00)00055-0 · Cited by 25 · see citing works
Asymmetric hydrogen transfer protocol for a synthesis of (+)-frontalin and (−)-malyngolide
Kanada, R. M., Taniguchi, T., & Ogasawara, K. (2000) · Tetrahedron Letters · doi:10.1016/s0040-4039(00)00430-5 · Cited by 22 · see citing works
Pyrimidine-based inhibitors of CaMKIIδ
Mavunkel, B. J., Xu, Y., Goyal, B., et al. (2008) · Bioorganic & Medicinal Chemistry Letters · doi:10.1016/j.bmcl.2008.02.056 · Cited by 20 · see citing works
PdII‐Catalyzed Spiroketalization of Ketoallylic Diols
Palmes, J. A., Paioti, P. H. S., de Souza, L. P., et al. (2013) · Chemistry - A European Journal · doi:10.1002/chem.201301723 · Cited by 20 · see citing works
Reaction Prediction and Synthesis Design
Goodman, J. M. (2018) · doi:10.1002/9783527806539.ch4b · Cited by 4 · see citing works
Learning to Plan Chemical SynthesesPreprint
Segler, M. H. S., Preuss, M., & Waller, M. P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1708.04202 · Cited by 4 · see citing works
Krueger, T., Vorndran, K., & Linker, T. (2010) · ChemInform · doi:10.1002/chin.201010079 · Cited by 2 · see citing works
ChemInform Abstract: Facile Assembly of Fused Benzo[4,5]furo Heterocycles.
Liu, J., Fitzgerald, A. E., & Mani, N. S. (2008) · ChemInform · doi:10.1002/chin.200833162 · Cited by 2 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
A Deep Learning Approach to Antibiotic Discovery
Stokes, J. M., Yang, K., Swanson, K., Jin, W., Cubillos-Ruiz, A., Donghia, N. M., MacNair, C. R., French, S., Carfrae, L. A., Bloom-Ackermann, Z., et al. (2020) · Cell · doi:10.1016/j.cell.2020.01.021 · Cited by 2,463 · see citing works
Artificial intelligence: A powerful paradigm for scientific research
Xu, Y., Liu, X., Cao, X., Huang, C., Liu, E., Qian, S., Liu, X., Wu, Y., Dong, F., Qiu, C.-W., et al. (2021) · The Innovation · doi:10.1016/j.xinn.2021.100179 · Cited by 1,726 · see citing works
Muratov, E. N., Bajorath, J., Sheridan, R. P., Tetko, I. V., Filimonov, D., Poroikov, V., Oprea, T. I., Baskin, I. I., Varnek, A., Roitberg, A., et al. (2020) · Chemical Society Reviews · doi:10.1039/d0cs00098a · Cited by 913 · see citing works
Artificial intelligence in chemistry and drug design
Brown, N., Ertl, P., Lewis, R., Luksch, T., Reker, D., & Schneider, N. (2020) · Journal of Computer-Aided Molecular Design · doi:10.1007/s10822-020-00317-x · Cited by 165 · see citing works
Application of machine learning for advanced material prediction and design
Chan, C. H., Sun, M., & Huang, B. (2022) · EcoMat · doi:10.1002/eom2.12194 · Cited by 123 · see citing works
Maley, S. M., Kwon, D.-H., Rollins, N., Stanley, J. C., Sydora, O. L., Bischof, S. M., & Ess, D. H. (2020) · Chemical Science · doi:10.1039/d0sc03552a · Cited by 81 · see citing works
Siebert, M., Krennrich, G., Seibicke, M., Siegle, A. F., & Trapp, O. (2019) · Chemical Science · doi:10.1039/c9sc04591k · Cited by 32 · see citing works
- Authors
- Gao, H., Struble, T. J., Coley, C. W., Wang, Y., Green, W. H., & Jensen, K. F.
- Affiliations
- Massachusetts Institute of Technology (US)
- Corresponding author
- Jensen, Klavs F.
- Venue
- ACS Central Science
- Publisher
- American Chemical Society (ACS)
- Published
- 2018-11-16
- Volume
- 4
- Issue
- 11
- Pages
- 1465-1476
- ISSN
- 2374-7943 (ISSN-L)
- 2374-7951
- PMC
- PMC6276053
- OpenAlex
- W2901942917
- Cited by
- 431
- References
- 48
- Citations by year
- 2026: 39 · 2025: 67 · 2024: 72 · 2023: 65 · 2022: 49 · 2021: 57 · 2020: 53 · 2019: 29
- FWCI
- 12.66
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- Locations
- ACS Central Science (publishedVersion, other-oa)
- Europe PMC (PubMed Central) (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Rare & Special e-Zone (The Hong Kong University of Science and Technology) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc)
- Funders
- Defense Advanced Research Projects Agency (ARO W911NF-16- 2-0023)
- Topics
- Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
- Physical Sciences › Chemistry › Physical and Theoretical Chemistry › History and advancements in chemistry
- Keywords
- organic synthesis, artificial neural networks, Reaxys, functional similarity
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
