A machine-learning tool to predict substrate-adaptive conditions for Pd-catalyzed C–N couplings | Science

A machine-learning tool to predict substrate-adaptive conditions for Pd-catalyzed C–N couplings | Science


31 Aug 2023

Vol 381, Issue 6661

pp. 965972

Editor’s summary

The palladium-catalyzed coupling of amines with aryl halides is one of the most widely used reactions in pharmaceutical research and manufacturing. Nonetheless, it depends sensitively on the structure of the two coupling partners and therefore often requires a trial-and-error process to identify pertinent optimal conditions. Rinehart et al. trained and validated a machine learning model to predict appropriate ligand, solvent, and base for coupling of particular reactant pairs. Ten products were isolated in more than 85% yield under the individualized conditions predicted by the model. —Jake S. Yeston


Machine-learning methods have great potential to accelerate the identification of reaction conditions for chemical transformations. A tool that gives substrate-adaptive conditions for palladium (Pd)–catalyzed carbon-nitrogen (C–N) couplings is presented. The design and construction of this tool required the generation of an experimental dataset that explores a diverse network of reactant pairings across a set of reaction conditions. A large scope of C–N couplings was actively learned by neural network models by using a systematic process to design experiments. The models showed good performance in experimental validation: Ten products were isolated in more than 85% yield from a range of couplings with out-of-sample reactants designed to challenge the models. Importantly, the developed workflow continually improves the prediction capability of the tool as the corpus of data grows.

Get full access to this article

View all available purchase options and get full access to this article.

Supplementary Materials

This PDF file includes:

Materials and Methods

Supplementary Text

Figs. S1 to S22

Tables S1 to S4

Scheme S1

Charts S1 and S2

NMR Spectra

References (3496)

References and Notes


A. S. Guram, R. A. Rennels, S. L. Buchwald, A Simple catalytic method for the conversion of aryl bromides to arylamines. Angew. Chem. Int. Ed.341348–1350 (1995).


J. Louie, J. F. Hartwig, Palladium-catalyzed synthesis of arylamines from aryl halides. Mechanistic studies lead to coupling in the absence of tin reagents. Tetrahedron Lett.363609–3612 (1995).


P. Ruiz-Castillo, S. L. Buchwald, Applications of palladium-catalyzed C–N cross-coupling reactions. Chem. Rev.11612564–12649 (2016).


J. F. Hartwig, K. H. Shaughnessy, S. Shekhar, R. A. Green, “Palladium-catalyzed amination of aryl halides” in vol. 100 of Organic ReactionsS. E. Denmark, Ed. (Wiley, 2019), pp. 853–958.


R. J. Lundgren, M. Stradiotto, Addressing challenges in palladium-catalyzed cross-coupling reactions through ligand design. Chemistry189758–9769 (2012).


B. T. Ingoglia, C. C. Wagen, S. L. Buchwald, Biaryl monophosphine ligands in palladium-catalyzed C–N coupling: An updated user’s guide. Tetrahedron754199–4211 (2019).


M. Fitzner, G. Wuitschik, R. J. Koller, J.-M. Adam, T. Schindler, J.-L. Reymond, What can reaction databases teach us about Buchwald-Hartwig cross-couplings? Chem. Sci.1113085–13093 (2020).


D. S. Surry, S. L. Buchwald, Dialkylbiaryl phosphines in Pd-catalyzed amination: A user’s guide. Chem. Sci.227–50 (2011).


M. D. Charles, P. Schultz, S. L. Buchwald, Efficient pd-catalyzed amination of heteroaryl halides. Org. Lett.73965–3968 (2005).


D. Maiti, B. P. Fors, J. L. Henderson, Y. Nakamura, S. L. Buchwald, Palladium-catalyzed coupling of functionalized primary and secondary amines with aryl and heteroaryl halides: Two ligands suffice in most cases. Chem. Sci.257–68 (2011).


E. P. K. Olsen, P. L. Arrechea, S. L. Buchwald, Mechanistic insight leads to a ligand which facilitates the palladium-catalyzed formation of 2-(hetero)arylaminooxazoles and 4-(hetero)arylaminothiazoles. Angew. Chem. Int. Ed.5610569–10572 (2017).


A. C. Sather, T. A. Martinot, Data-rich experimentation enables palladium-catalyzed couplings of piperidines and five-membered (hetero)aromatic electrophiles. Org. Process Res. Dev.231725–1739 (2019).


B. J. Shields, J. Stevens, J. Li, M. Parasram, F. Damani, J. I. M. Alvarado, J. M. Janey, R. P. Adams, A. G. Doyle, Bayesian reaction optimization as a tool for chemical synthesis. Nature59089–96 (2021).


F. Häse, L. M. Roch, C. Kreisbeck, A. Aspuru-Guzik, Phoenics: A Bayesian optimizer for chemistry. ACS Cent. Sci.41134–1145 (2018).


D. T. Ahneman, J. G. Estrada, S. Lin, S. D. Dreher, A. G. Doyle, Predicting reaction performance in C–N cross-coupling using machine learning. Science360186–190 (2018).


K. D. Collins, F. Glorius, Intermolecular reaction screening as a tool for reaction evaluation. Acc. Chem. Res.48619–627 (2015).


P. Schwaller, A. C. Vaucher, T. Laino, J.-L. Reymond, Prediction of chemical reaction yields using deep learning. Mach. Learn. Sci. Technol.2015016 (2021).


T. R. Gimadiev, A. Lin, V. A. Afonina, D. Batyrshin, R. I. Nugmanov, T. Akhmetshin, P. Sidorov, N. Duybankova, J. Verhoeven, J. Wegner, H. Ceulemans, A. Gedich, T. I. Madzhidov, A. Varnek, Reaction data curation I: Chemical structures and transformations standardization. Mol. Inform.40e2100119 (2021).


J. Li, M. D. Eastgate, Making better decisions during synthetic route design: Leveraging prediction to achieve greenness-by-design. React. Chem. Eng.41595–1607 (2019).


M. Fitzner, G. Wuitschik, R. Koller, J.-M. Adam, T. Schindler, Machine learning C–N couplings: Obstacles for a general-purpose reaction yield prediction. ACS Omega83017–3025 (2023).


N. I. Rinehart, R. K. Saunthwal, J. Wellauer, A. F. Zahrt, L. Schlemper, A. S. Shved, R. Bigler, S. Fantasia, S. E. Denmark, Development and validation of a chemoinformatic workflow for predicting reaction yield for Pd-catalyzed C–N couplings with substrate generalizability. chemRxiv chemrixiv-2022-hspvw-v2 [Preprint] (2023); https://doi.org/10.26434/chemrxiv-2022-hspwv-v2.


E. Shim, J. A. Kammeraad, Z. Xu, A. Tewari, T. Cernak, P. M. Zimmerman, Predicting reaction conditions from limited data through active transfer learning. Chem. Sci.136655–6668 (2022).


N. H. Angello, V. Rathore, W. Beker, A. Wołos, E. R. Jira, R. Roszak, T. C. Wu, C. M. Schroeder, A. Aspuru-Guzik, B. A. Grzybowski, M. D. Burke, Closed-loop optimization of general reaction conditions for heteroaryl Suzuki-Miyaura coupling. Science378399–405 (2022).



M. K. Nielsen, D. T. Ahneman, O. Riera, A. G. Doyle, Deoxyfluorination with sulfonyl fluorides: Navigating reaction space with machine learning. J. Am. Chem. Soc.1405004–5008 (2018).


M. Yamaguchi, K. Suzuki, Y. Sato, K. Manabe, Palladium-catalyzed direct C3-selective arylation of N-unsubstituted indoles with aryl chlorides and triflates. Org. Lett.195388–5391 (2017).


S. Basak, S. Dutta, D. Maiti, Accessing C2-functionalized 1,3-(benz)azoles through transition metal-catalyzed C–H activation. Chemistry2710533–10557 (2021).


D. S. Surry, S. L. Buchwald, Biaryl phosphane ligands in palladium-catalyzed amination. Angew. Chem. Int. Ed.476338–6361 (2008).


H. Chen, M. Lei, L. Hu, Synthesis of 1-aryl indoles via coupling reaction of indoles and aryl halides catalyzed by CuI/metformin. Tetrahedron705626–5631 (2014).


S. A. Iqbal, K. Yuan, J. Cid, J. Pahl, M. J. Ingleson, Controlling selectivity in N-heterocycle directed borylation of indoles. Org. Biomol. Chem.192949–2958 (2021).


J.-K. Kwon, J.-H. Lee, T. S. Kim, E. K. Yum, H. J. Park, Diversification of indoles via microwave-assisted ligand-free copper-catalyzed N-arylation. Bull. Korean Chem. Soc.371927–1933 (2016).


A. Shoberu, C.-K. Li, H.-F. Qian, J.-P. Zou, Copper-catalyzed, N-auxiliary group-controlled switchable transannulation/nitration initiated by nitro radicals: Selective synthesis of pyridoquinazolones and 3-nitroindoles. Org. Chem. Front.85821–5830 (2021).


A. Gangjee, R. Devraj, F.-T. Lin, Synthesis of 2,4-diamino-5,10-dideaza nonclassical antifolates. J. Heterocycl. Chem.281747–1751 (1991).


A. N. Butkevich, V. N. Belov, K. Kolmakov, V. V. Sokolov, H. Shojaei, S. C. Sidenstein, D. Kamin, J. Matthias, R. Vlijm, J. Engelhardt, S. W. Hell, Hydroxylated fluorescent dyes for live-cell labeling: Synthesis, spectra and super-resolution STED. Chemistry2312114–12119 (2017).


H.-G. Cheng, L.-Q. Lu, T. Wang, Q.-Q. Yang, X.-P. Liu, Y. Li, Q.-H. Deng, J.-R. Chen, W.-J. Xiao, Highly enantioselective Friedel-Crafts alkylation/N-hemiacetalization cascade reaction with indoles. Angew. Chem. Int. Ed.523250–3254 (2013).


M. E. Muratore, C. A. Holloway, A. W. Pilling, R. I. Storer, G. Trevitt, D. J. Dixon, Enantioselective Brønsted acid-catalyzed N-acyliminium cyclization cascades. J. Am. Chem. Soc.13110796–10797 (2009).


M. Miyashita, A. Yoshikoshi, P. A. Grieco, Pyridinium p-toluenesulfonate. A mild and efficient catalyst for the tetrahydropyranylation of alcohols. J. Org. Chem.423772–3774 (1977).


S. J. Barraza, S. E. Denmark, Synthesis, reactivity, functionalization, and ADMET properties of silicon-containing nitrogen heterocycles. J. Am. Chem. Soc.1406668–6684 (2018).


A. L. Johnson, W. A. Price, P. C. Wong, R. F. Vavala, J. M. Stump, Synthesis and pharmacology of the potent angiotensin-converting enzyme inhibitor N-[1([1(S)-(ethoxycarbonyl)-3-phenylpropyl]-(S)-alanyl-(S)-pyroglutamic acid. J. Med. Chem.281596–1602 (1985).


A. F. Zahrt, J. J. Henle, S. E. Denmark, Cautionary guidelines for machine learning studies with combinatorial datasets. ACS Comb. Sci.22586–591 (2020).


[ PubMed ][ Cross Ref ]Virtanen P., Gumers R., TE Oliphant, M. Haberland, T. Reddy, D. Cournapeau, E. Burovski, P. Peterson, W. Weckesser, J. Bright, SJ van der Walt, M. Brett, J.S. Wilson , Millman KJ , Mayorov N , Nelson RJ , Jones E , Kern R , Larson E , Carey CJ , İ. Polat , Y Feng , EW Moore , J VanderPlas , D Laxalde , J Perktold , R Cimrman , I Henriksen , EA Quintero , CR Harris , AM Archibald , AH Ribeiro , F Pedregosa , P van Mulbregt ; SciPy 1.0 Contributors, SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python. Nat. Methods17261–272 (2020).


CR Harris, KJ Millman, SJ van der Walt, R. Gommers, P. Virtanen, D. Cournapeau, E. Wieser, J. Taylor, S. Berg, NJ Smith, R. Kern, M. Picus, S. Hoyer, MH van Kerkwijk, M. Brett, A. Haldane, JF Del Río, M. Wiebe, P. Peterson, P. Gérard-Marchant, K. Sheppard, T. Reddy, W. Weckesser, H. Abbasi, C. Gohlke, TE Oliphant, Array programming with NumPy. Nature585357–362 (2020).


F Pedregosa , G Varoquaux , A Gramfort , V Michel , B Thirion , O Grisel , M Blondel , P Prettenhofer , R Weiss , V Dubourg , J Vanderplas , A Passos , D . Cornaskin, M. Brucher, M. Dog, É. Duchesnay, Scikit-Learn: Machine Learning in Python. J. Mach. Learn. Res.122825–2830 (2011).


J. J. Henle, A. F. Zahrt, B. T. Rose, W. T. Darrow, Y. Wang, S. E. Denmark, Development of a computer-guided workflow for catalyst optimization. Descriptor validation, subset selection, and training set analysis. J. Am. Chem. Soc.14211578–11592 (2020).


A. F. Zahrt, J. J. Henle, B. T. Rose, Y. Wang, W. T. Darrow, S. E. Denmark, Prediction of higher-selectivity catalysts by computer-driven workflow and machine learning. Science363eaau5631 (2019).


S. Grimme, Exploration of chemical compound, conformer, and reaction space with meta-dynamics simulations based on tight-binding quantum chemical calculations. J. Chem. Theory Comput.152847–2862 (2019).


Bannwarth C, Caldeweyher E, Ehlert S, Hansen A, Pracht P, Seibert J, Spicher S, Grimme S. Extended tight-binding quantum chemistry methods. Wiley Interdiscip. Rev. Comput. Mol. Sci.11e1493 (2021).


K. V. Chuang, M. J. Keiser, Comment on “Predicting reaction performance in C–N cross-coupling using machine learning”. Science362eaat8603 (2018).


J. G. Estrada, D. T. Ahneman, R. P. Sheridan, S. D. Dreher, A. G. Doyle, Response to comment on “Predicting reaction performance in C–N cross-coupling using machine learning”. Science362eaat8763 (2018).


As a possible explanation, some products may form exciplexes at concentrations reached during high-performance liquid chromatography (HPLC) separation and ultraviolet (UV) detection. This effect should lead to nonlinearity between product-to-standard mol ratio and absorbance ratios with a lower measured absorbance than a linear function would predict at higher product concentrations. In practice, most products in this work showed some nonlinearity, and a second-order polynomial was used for calibration curves.


A. J. DeAngelis, P. G. Gildner, R. Chow, T. J. Colacot, Generating active “L-Pd(0)” via neutral or cationic π-allylpalladium complexes featuring biaryl/bipyrazolylphosphines: Synthetic, mechanistic, and structure-activity studies in challenging cross-coupling reactions. J. Org. Chem.806794–6813 (2015).


N. C. Bruno, M. T. Tudge, S. L. Buchwald, Design and preparation of new palladium precatalysts for C–C and C–N cross-coupling reactions. Chem. Sci.4916–920 (2013).



G. A. Grasa, M. S. Viciu, J. Huang, S. P. Nolan, Amination reactions of aryl halides with nitrogen-containing reagents mediated by palladium/imidazolium salt systems. J. Org. Chem.667729–7737 (2001).


S. Ueda, S. L. Buchwald, Catalyst-controlled chemoselective arylation of 2-aminobenzimidazoles. Angew. Chem. Int. Ed.5110364–10367 (2012).


P. Ruiz-Castillo, D. G. Blackmond, S. L. Buchwald, Rational ligand design for the arylation of hindered primary amines guided by reaction progress kinetic analysis. J. Am. Chem. Soc.1373085–3092 (2015).


C. Yuan, L. Zhang, Y. Zhao, Cu(II)-catalyzed C–N coupling of (hetero)aryl halides and N-nucleophiles promoted by α-benzoin oxime. Molecules244177 (2019).


B. P. Fors, K. Dooleweerdt, Q. Zeng, S. L. Buchwald, An efficient system for the Pd-catalyzed cross-coupling of amides and aryl chlorides. Tetrahedron656576–6583 (2009).


A. Gangjee, O. A. Namjoshi, S. Raghavan, S. F. Queener, R. L. Kisliuk, V. Cody, Design, synthesis, and molecular modeling of novel pyrido[2,3-d]pyrimidine analogues as antifolates; Application of Buchwald-Hartwig aminations of heterocycles. J. Med. Chem.564422–4441 (2013).


G. Song, L. Yang, J.-S. Li, W.-J. Tang, W. Zhang, R. Cao, C. Wang, J. Xiao, D. Xue, Chiral arylated amines via C–N coupling of chiral amines with aryl bromides promoted by light. Angew. Chem. Int. Ed.6021536–21542 (2021).


J. Duan, P. Y. Choy, K. B. Gan, F. Y. Kwong, N-Difluoromethylation of N-pyridyl-substituted anilines with ethyl bromodifluoroacetate. Org. Biomol. Chem.201883–1887 (2022).


J. Shao, X. Huang, S. Wang, B. Liu, B. Xu, A straightforward synthesis of N-monosubstituted α-keto amides via aerobic benzylic oxidation of amides. Tetrahedron68573–579 (2012).


V. Soni, U. N. Patel, B. Punji, Metal-free regioselective C-3 acetoxylation of N-substituted indoles: Crucial impact of nitrogensubstituent. RSC Advances557472–57481 (2015).


R. J. Key, A. K. Vannucci, Nickel dual photoredox catalysis for the synthesis of aryl amines. Organometallics371468–1472 (2018).


C. A. Malapit, N. Ichiishi, M. S. Sanford, Pd-Catalyzed decarbonylative cross-couplings of aroyl chlorides. Org. Lett.194142–4145 (2017).


Z. Chen, H. Zeng, S. A. Girard, F. Wang, N. Chen, C.-J. Li, Formal direct cross-coupling of phenols with amines. Angew. Chem. Int. Ed.5414487–14491 (2015).


M. O. Akram, A. Das, I. Chakrabarty, N. T. Patil, Ligand-enabled gold-catalyzed C(sp2)-N cross-coupling reactions of aryl iodides with amines. Org. Lett.218101–8105 (2019).


G. Manolikakes, A. Gavryushin, P. Knochel, An efficient silane-promoted nickel-catalyzed amination of aryl and heteroaryl chlorides. J. Org. Chem.731429–1434 (2008).


M. P. Groziak, S. R. Wilson, G. L. Clauson, N. J. Leonard, Fluorescent heterocyclic systems: Syntheses, structures, and physicochemical properties of dipyrido-substituted 1,3,4,6-tetraazapentalenes. J. Am. Chem. Soc.1088002–8006 (1986).


K. Saito, H. Miyashita, T. Akiyama, Asymmetric transfer hydrogenation of ketimines by indoline as recyclable hydrogen donor. Org. Lett.165312–5315 (2014).


T. Akiyama, K. Saito, S. Janczak, Z. J. Lesnikowski, Phosphoric acid bridged cobalt bis(dicarbollide) ion as a highly efficient catalyst for the organocatalytic hydrogenation of ketimines. Synlett25795–798 (2014).


A. Mital, D. Murugesan, M. Kaiser, C. Yeates, I. H. Gilbert, Discovery and optimisation studies of antimalarial phenotypic hits. Eur. J.Med. Chem.103530–538 (2015).


D. Murugesan, A. Mital, M. Kaiser, D. M. Shackleford, J. Morizzi, K. Katneni, M. Campbell, A. Hudson, S. A. Charman, C. Yeates, I. H. Gilbert, Discovery and structure-activity relationships of pyrrolone antimalarials. J. Med. Chem.562975–2990 (2013).


NV Shevchuk, K. Liubchak, KG Nazarenko, AA Yurchenko, DM Volochnyuk, OO Grygorenko, AA Tolmachev, A convenient synthesis of (1).H-azol-1-yl)piperidines. Synthesis442041–2048 (2012).


C. E. Coomber, V. Laserna, L. T. Martin, P. D. Smith, H. C. Hailes, M. J. Porter, T. D. Sheppard, Catalytic direct amidations in tert-butyl acetate using B(OCH2CF3)3. Org. Biomol. Chem.176465–6469 (2019).


T. N. Ansari, A. Taussat, A. H. Clark, M. Nachtegaal, S. Plummer, F. Gallou, S. Handa, Insights on bimetallic micellar nanocatalysis for Buchwald–Hartwig aminations. ACS Catal.910389–10397 (2019).


Y. Zhao, S. Zhang, Y. Yamamoto, M. Bao, T. Jin, M. Terada, Heterogeneous catalytic reduction of tertiary amides with hydrosilanes using unsupported nanoporous gold catalyst. Adv. Synth. Catal.3614817–4824 (2019).


KP Jayasundera, TGW Engels, DJ Lun, Mungalpara MN, Plieger PG, Rowlands GJ, The synthesis of planar chiral pseudo-gem aminophosphine pre-ligands based on [2.2]paracyclophane. Org. Biomol. Chem.158975–8984 (2017).


L. Alakonda, M. Periasamy, Simple and convenient methods for N-arylation of heterocycles and diphenylamine. Synthesis441063–1068 (2012).


D. Cheng, F. Gan, W. Qian, W. Bao, d-Glucosamine—A natural ligand for the N-arylation of imidazoles with aryl and heteroaryl bromides catalyzed by CuI. Green Chem.10171–173 (2008).


B. Weng, J.-H. Li, Betti base-derived tetradentate ligand: Synthesis and application in copper-catalyzed N-arylation of imidazoles. Appl. Organomet. Chem.23375–378 (2009).


C.-H. Lim, M. Kudisch, B. Liu, G. M. Miyake, C–N cross-coupling via photoexcitation of nickel-amine complexes. J. Am. Chem. Soc.1407667–7673 (2018).


G. L. Beutner, J. R. Coombs, R. A. Green, B. Inankur, D. Lin, J. Qiu, F. Roberts, E. M. Simmons, S. R. Wisniewski, Palladium-catalyzed amid ation and amination of (hetero)aryl hlorides under homogeneous conditions enabled by a soluble DBU/NaTFA dual-base system. Org. Process Res. Dev.231529–1537 (2019).


B. R. Trowse, F. P. Byrne, J. Sherwood, P. O’Brien, J. Murray, T. Farmer, 2,2,5,5-Tetramethyloxolane (TMO) as a solvent for Buchwald–Hartwig aminations. ACS Sustain. Chem. Eng.917330–17337 (2021).


Chen Y, Du F, Chen F, Zhou Q, Chen G. Methyl-α-d-glucopyranoside as green ligand for selective copper-catalyzed N-arylation. Synthesis514590–4600 (2019).


X.-L. Zhang, R.-L. Guo, M.-Y. Wang, B.-Y. Zhao, Q. Jia, J.-H. Yang, Y.-Q. Wang, Palladium-catalyzed three-component regioselective dehydrogenative coupling of indoles, 2-methylbut-2-ene, and carboxylic acids. Org. Lett.239574–9579 (2021).



A. H. Sommers, S. E. Aaland, The reaction of 1,5-dibromopentane with o-substituted anilines. The synthesis of 1-arylpiperi dines. J. Am. Chem. Soc.755280–5283 (1953).


T. Mino, Y. Harada, H. Shindo, M. Sakamoto, T. Fujita, Copper-catalyzed N-arylation of amides and azoles using phosphine-free hydrazone ligands. Synlett2008614–620 (2008).


X. Yang, H. Xing, Y. Zhang, Y. Lai, Y. Zhang, Y. Jiang, D. Ma, CuI/8-hydroxyquinalidine promoted N-arylation of indole and azoles. Chin. J. Chem.30875–880 (2012).


J. D. Grayson, F. M. Dennis, C. C. Robertson, B. M. Partridge, Chan-Lam amination of secondary and tertiary benzylic boronic esters. J. Org. Chem.869883–9897 (2021).


V. Skrypai, S. E. Varjosaari, F. Azam, T. M. Gilbert, M. J. Adler, Chiral Brønsted acid-catalyzed metal-free asymmetric direct reductive amination using 1-hydrosilatrane. J. Org. Chem.845021–5026 (2019).


J. H. Pang, A. Kaga, S. Chiba, Nucleophilic amination of methoxypyridines by a sodium hydride-iodide composite. Chem. Commun.5410324–10327 (2018).

Information & Authors


Published In


Volume 381 | Issue 6661
1 September 2023


Copyright © 2023 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works.

Article versions

Submission history

Received: 8 December 2022

Accepted: 1 August 2023

Published in print: 1 September 2023


Request permissions for this article.


We acknowledge the support services of the NMR, mass spectrometry, and microanalytical laboratories of the University of Illinois at Urbana-Champaign.

Funding: We acknowledge generous financial support from Hoffmann La-Roche. A.S.S. thanks the National Science Foundation (NSF) (grant no. CHE 1900617) for financial support. This work was also supported in part by the Molecule Maker Lab Institute, an AI Research Institutes program supported by the NSF under grant no. CHE 2019897. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect those of the NSF.

Author contributions: N.I.R. contributed to the dataset, wrote the main code package, designed new descriptors, tested neural network architectures for modeling, conceptualized the informatics-guided strategy to exploring the reaction space as a network, designed and contributed to the validation experiments, contributed to composing the manuscript, and wrote the Computational and informatics methods, Experimental methods, Workflow development, General references, Experimental validation, and Master list of dataset products sections of the SM; R.K.S. contributed experimentally to the dataset, contributed to the validation experiments, made authentic products for zero-yielding plates, and composed most of the Characterization data, Characterization references, and NMR spectra sections of the SM; J.W. contributed experimental results to the dataset and prepared reference materials; A.F.Z. helped conceptualize the scope of the project and design descriptors as well as initial classification modeling efforts; L.S. prepared palladium complexes and reference materials; A.S.S. shared code that facilitated the calculation and testing of new descriptors; R.B. conceived of and supervised the project, designed the experimental part of the workflow, contributed experimental results to the dataset, and contributed to composing the manuscript; and both S.F. and S.E.D. conceptualized and supervised the project and contributed to composing the manuscript.

Competing interests: The authors declare no competing interests.

Data and materials availability: Full experimental procedures, including validation runs, characterization data, experimental apparatus, qHPLC analytical methodology, and copies of 1H, 13C, 31P, and 19F spectra as well as feature engineering, modeling details and model validation, structures of each product made in the dataset, and predictions for each condition with every reactant pair in the dataset can be found in the SM. Code developed in this work is available at Zenodo (33), and ongoing development through our laboratory’s Github organization will be available at github.com/SEDenmarkLab/Lucid_Somnambulist.



Roger Adams Laboratory, Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA.

Roles: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Validation, Visualization, Writing – original draft, and Writing – review & editing.

Roger Adams Laboratory, Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA.

Roles: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, and Writing – review & editing.

Joel Wellauer

Pharmaceutical Division, Synthetic Molecules Technical Development, Process Chemistry and Catalysis, F. Hoffmann–La Roche, Ltd., Basel, Switzerland.

Roles: Conceptualization, Formal analysis, Investigation, Methodology, Validation, and Visualization.

Roger Adams Laboratory, Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA.

Roles: Conceptualization, Formal analysis, Methodology, Project administration, Software, Supervision, and Writing – review & editing.

Lukas Schlemper

Pharmaceutical Division, Synthetic Molecules Technical Development, Process Chemistry and Catalysis, F. Hoffmann–La Roche, Ltd., Basel, Switzerland.

Roles: Investigation, Resources, and Validation.

Roger Adams Laboratory, Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA.

Role: Software.

Pharmaceutical Division, Synthetic Molecules Technical Development, Process Chemistry and Catalysis, F. Hoffmann–La Roche, Ltd., Basel, Switzerland.

Roles: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, and Writing – review & editing.

Pharmaceutical Division, Synthetic Molecules Technical Development, Process Chemistry and Catalysis, F. Hoffmann–La Roche, Ltd., Basel, Switzerland.

Roles: Conceptualization, Project administration, Supervision, Visualization, and Writing – review & editing.

Roger Adams Laboratory, Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA.

Roles: Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, and Writing – review & editing.

Funding Information

Molecule Maker Laboratory Institute: CHE 2019897


Metrics & Citations


Article Usage



Cite as

A machine-learning tool to predict substrate-adaptive conditions for Pd-catalyzed C–N couplings.Science381,965-972(2023).DOI:10.1126/science.adg2114

Export citation

Select the format you want to export the citation of this publication.

View Options

Check Access





Read More

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Most Popular

On Key

Related Posts

Traces, faces and route

Traces, faces and route

Rodrigo Muñoz León Hill of Paltacalo The journey to Paltacalo, approximately 4 hours—archaeological territory; It belongs to the province of El Oro, but has a