Skip to main content

Advertisement

SpringerLink
Log in

Machine learning-aided generative molecular design

  • Review Article
  • Published:

From Nature Machine Intelligence

View current issue Submit your manuscript

Abstract

Machine learning has provided a means to accelerate early-stage drug discovery by combining molecule generation and filtering steps in a single architecture that leverages the experience and design preferences of medicinal chemists. However, designing machine learning models that can achieve this on the fly to the satisfaction of medicinal chemists remains a challenge owing to the enormous search space. Researchers have addressed de novo design of molecules by decomposing the problem into a series of tasks determined by design criteria. Here we provide a comprehensive overview of the current state of the art in molecular design using machine learning models as well as important design decisions, such as the choice of molecular representations, generative methods and optimization strategies. Subsequently, we present a collection of practical applications in which the reviewed methodologies have been experimentally validated, encompassing both academic and industrial efforts. Finally, we draw attention to the theoretical, computational and empirical challenges in deploying generative machine learning and highlight future opportunities to better align such approaches to achieve realistic drug discovery end points.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1: Generative ML-assisted molecular design pipeline.
Fig. 2: Illustrations for generative tasks, generative strategies and molecular representations.
Fig. 3: Selected examples of experimentally validated generative designs.

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

References

  1. Delaney, J. S. ESOL: estimating aqueous solubility directly from molecular structure. J. Chem. Inf. Comput. 44, 1000–1005 (2004).

    Google Scholar 

  2. Gillette, J. R., Mitchell, J. R. & Brodie, B. B. Biochemical mechanisms of drug toxicity. Annu. Rev. Pharmacol. 14, 271–288 (1974).

    Google Scholar 

  3. Gibaldi, M. & Perrier, D. Pharmacokinetics (CRC Press, 1982).

  4. Bohacek, R. S., McMartin, C. & Guida, W. C. The art and practice of structure-based drug design: a molecular modeling perspective. Med. Res. Rev. 16, 3–50 (1996).

    Google Scholar 

  5. Stumpfe, D. & Bajorath, J. Exploring activity cliffs in medicinal chemistry: miniperspective. J. Med. Chem. 55, 2932–2942 (2012).

    Google Scholar 

  6. Scannell, J. W., Blanckley, A., Boldon, H. & Warrington, B. Diagnosing the decline in pharmaceutical R&D efficiency. Nat. Rev. Drug Discov. 11, 191–200 (2012).

    Google Scholar 

  7. Berdigaliyev, N. & Aljofan, M. An overview of drug discovery and development. Future Med. Chem. 12, 939–947 (2020).

    Google Scholar 

  8. Ringel, M. S., Scannell, J. W., Baedeker, M. & Schulze, U. Breaking Eroom’s law. Nat. Rev. Drug Discov. 19, 833–834 (2020).

    Google Scholar 

  9. Aparoy, P., Kumar Reddy, K. & Reddanna, P. Structure and ligand based drug design strategies in the development of novel 5-LOX inhibitors. Curr. Med. Chem. 19, 3763–3778 (2012).

    Google Scholar 

  10. Baskin, I. & Varnek, A. in Chemoinformatics Approaches to Virtual Screening Ch. 1, 1–43 (Royal Society of Chemistry, 2008).

  11. Kuntz, I. D. Structure-based strategies for drug design and discovery. Science 257, 1078–1082 (1992).

    Google Scholar 

  12. Anderson, A. C. The process of structure-based drug design. Chem. Biol. 10, 787–797 (2003).

    Google Scholar 

  13. Choung, O.-H., Vianello, R., Segler, M., Stiefl, N. & Jiménez-Luna, J. Extracting medicinal chemistry intuition via preference machine learning. Nat. Commun. 14, 6651 (2023).

    Google Scholar 

  14. Lyu, J. et al. Ultra-large library docking for discovering new chemotypes. Nature 566, 224–229 (2019). Ultralarge-scale virtual screening of a make-on-demand library identified hits with previously unknown chemical motifs that were experimentally validated.

    Google Scholar 

  15. Sadybekov, A. A. et al. Synthon-based ligand discovery in virtual libraries of over 11 billion compounds. Nature 601, 452–459 (2022).

    Google Scholar 

  16. Gorgulla, C. et al. Virtualflow 2.0—the next generation drug discovery platform enabling adaptive screens of 69 billion molecules. Preprint at bioRxiv https://doi.org/10.1101/2023.04.25.537981 (2023).

  17. Rombach, R., Blattmann, A., Lorenz, D., Esser, P. & Ommer, B. High-resolution image synthesis with latent diffusion models. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition 10684–10695 (IEEE, 2022).

  18. Brown, T. et al. Language models are few-shot learners. Adv. Neural Inf. Process. Syst. 33, 1877–1901 (2020).

  19. Zhavoronkov, A. et al. Deep learning enables rapid identification of potent DDR1 kinase inhibitors. Nat. Biotechnol. 37, 1038–1040 (2019). One of the first studies to experimentally validate ML-generated molecules and highlighted the potential for accelerated drug discovery.

    Google Scholar 

  20. Ren, F. et al. AlphaFold accelerates artificial intelligence powered drug discovery: efficient discovery of a novel CDK20 small molecule inhibitor. Chem. Sci. 14, 1443–1452 (2023). A study performing molecular docking using an AlphaFold-generated structure on ML-generated molecules with experimental validation.

    Google Scholar 

  21. Wu, C.-T. et al. COT: an efficient and accurate method for detecting marker genes among many subtypes. Bioinf. Adv. 2, vbac037 (2022).

    Google Scholar 

  22. Méndez-Lucio, O., Baillif, B., Clevert, D.-A., Rouquié, D. & Wichard, J. De novo generation of hit-like molecules from gene expression signatures using artificial intelligence. Nat. Commun. 11, 10 (2020).

    Google Scholar 

  23. Sanchez-Fernandez, A., Rumetshofer, E., Hochreiter, S. & Klambauer, G. CLOOME: contrastive learning unlocks bioimaging databases for queries with chemical structures. Nat. Commun. 14, 7339 (2023).

    Google Scholar 

  24. Nguyen, C. Q., Pertusi, D. & Branson, K. M. Molecule-morphology contrastive pretraining for transferable molecular representation. Preprint at https://arxiv.org/abs/2305.09790 (2023).

  25. Schaller, D. et al. Next generation 3D pharmacophore modeling. Wiley Interdiscip. Rev. Comput. Mol. Sci. 10, e1468 (2020).

    Google Scholar 

  26. Imrie, F., Hadfield, T. E., Bradley, A. R. & Deane, C. M. Deep generative design with 3D pharmacophoric constraints. Chem. Sci. 12, 14577–14589 (2021).

    Google Scholar 

  27. Guo, J. et al. Link-INVENT: generative linker design with reinforcement learning. Digit. Discov. 2, 392–408 (2023).

    Google Scholar 

  28. Torge, J., Harris, C., Mathis, S. V. & Lio, P. DiffHopp: a graph diffusion model for novel drug design via scaffold hopping. Preprint at https://arxiv.org/abs/2308.07416 (2023).

  29. Keiser, M. J. et al. Relating protein pharmacology by ligand chemistry. Nat. Biotechnol. 25, 197–206 (2007).

    Google Scholar 

  30. Harris, C. et al. Flexible small-molecule design and optimization with equivariant diffusion models. In ICLR 2023—Machine Learning for Drug Discovery Workshop (OpenReview, 2023).

  31. Weininger, D. SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules. J. Chem. Inf. Comput. Sci. 28, 31–36 (1988).

    Google Scholar 

  32. Krenn, M., Häse, F., Nigam, A., Friederich, P. & Aspuru-Guzik, A. Self-referencing embedded strings (SELFIES): a 100% robust molecular string representation. Mach. Learn. Sci. Technol. 1, 045024 (2020).

    Google Scholar 

  33. Kingma, D. P. & Welling, M. Auto-encoding variational bayes. In Proc. 2nd International Conference on Learning Representations (eds Bengio, Y. & LeCun, Y.) (OpenReview, 2014).

  34. Gómez-Bombarelli, R. et al. Automatic chemical design using a data-driven continuous representation of molecules. ACS Cent. Sci. 4, 268–276 (2018). One of the first studies to apply a variational autoencoder for molecular design and is a foundational work for many recently reported methods.

    Google Scholar 

  35. Simonovsky, M. & Komodakis, N. GraphVAE: towards generation of small graphs using variational autoencoders. In Artificial Neural Networks and Machine Learning—ICANN 2018 (eds Kůrková, V. et al.) 412–422 (Springer, 2018).

  36. Goodfellow, I. et al. Generative adversarial nets. Adv. Neural Inf. Process. Syst. 27, 2672–2680 (2014).

    Google Scholar 

  37. Guimaraes, G. L., Sanchez-Lengeling, B., Outeiral, C., Farias, P. L. C. & Aspuru-Guzik, A. Objective-reinforced generative adversarial networks (ORGAN) for sequence generation models. Preprint at https://arxiv.org/abs/1705.10843 (2017).

  38. De Cao, N. & Kipf, T. MolGAN: an implicit generative model for small molecular graphs. Preprint at https://arxiv.org/abs/1805.11973 (2018).

  39. Rezende, D. & Mohamed, S. Variational inference with normalizing flows. In Proc. 32nd International Conference on Machine Learning 1530–1538 (PMLR, 2015).

  40. Shi, C. et al. GraphAF: a flow-based autoregressive model for molecular graph generation. In Proc. 8th International Conference on Learning Representations (OpenReview, 2020).

  41. Lipman, Y., Chen, R. T., Ben-Hamu, H., Nickel, M. & Le, M. Flow matching for generative modeling. In Proc. 11th International Conference on Learning Representations (OpenReview, 2023).

  42. Song, Y. et al. Equivariant flow matching with hybrid probability transport for 3D molecule generation. Adv. Neural Inf. Process. Syst. 36, 549–568 (2023).

  43. Van Oord, A., Kalchbrenner, N. & Kavukcuoglu, K. Pixel recurrent neural networks. In Proc. 33rd International Conference on Machine Learning 1747–1756 (PMLR, 2016).

  44. Popova, M., Shvets, M., Oliva, J. & Isayev, O. MolecularRNN: generating realistic molecular graphs with optimized properties. Preprint at https://arxiv.org/abs/1905.13372 (2019).

  45. Gebauer, N., Gastegger, M. & Schütt, K. Symmetry-adapted generation of 3D point sets for the targeted discovery of molecules. Adv. Neural Inf. Process. Syst. 32, 7566–7578 (2019).

    Google Scholar 

  46. Ho, J., Jain, A. & Abbeel, P. Denoising diffusion probabilistic models. Adv. Neural Inf. Process. Syst. 33, 6840–6851 (2020).

  47. Hoogeboom, E., Satorras, V. G., Vignac, C. & Welling, M. Equivariant diffusion for molecule generation in 3D. In Proc. 39th International Conference on Machine Learning 8867–8887 (PMLR, 2022).

  48. Schneuing, A. et al. Structure-based drug design with equivariant diffusion models. In NeurIPS 2022 Machine Learning for Structural Biology (OpenReview, 2022). One of the first studies to leverage the flexibility of diffusion models to achieve a variety of types of conditional generation and molecule optimization.

  49. Igashov, I. et al. Equivariant 3D-conditional diffusion models for molecular linker design. Nat. Mach. Intell. 6, 417–427 (2024).

    Google Scholar 

  50. Xu, M. et al. GeoDiff: a geometric diffusion model for molecular conformation generation. In Proc. 9th International Conference on Learning Representations (OpenReview, 2021).

  51. Liu, Q., Allamanis, M., Brockschmidt, M. & Gaunt, A. Constrained graph variational autoencoders for molecule design. Adv. Neural Inf. Process. Syst. 31, 7795–7804 (2018).

    Google Scholar 

  52. Segler, M. H., Kogej, T., Tyrchan, C. & Waller, M. P. Generating focused molecule libraries for drug discovery with recurrent neural networks. ACS Cent. Sci. 4, 120–131 (2018).

    Google Scholar 

  53. Bengio, E., Jain, M., Korablyov, M., Precup, D. & Bengio, Y. Flow network based generative models for non-iterative diverse candidate generation. Adv. Neural Inf. Process. Syst. 34, 27381–27394 (2021).

    Google Scholar 

  54. Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).

    Google Scholar 

  55. Jensen, J. H. A graph-based genetic algorithm and generative model/Monte Carlo tree search for the exploration of chemical space. Chem. Sci. 10, 3567–3572 (2019).

    Google Scholar 

  56. Yang, X., Zhang, J., Yoshizoe, K., Terayama, K. & Tsuda, K. ChemTS: an efficient Python library for de novo molecular generation. Sci. Technol. Adv. Mat. 18, 972–976 (2017).

    Google Scholar 

  57. Olivecrona, M., Blaschke, T., Engkvist, O. & Chen, H. Molecular de-novo design through deep reinforcement learning. J. Cheminf. 9, 48 (2017). One of the first studies applying reinforcement learning to molecular design and is the first version of ‘REINVENT’, an industrially used method, which is still under active development.

    Google Scholar 

  58. Popova, M., Isayev, O. & Tropsha, A. Deep reinforcement learning for de novo drug design. Sci. Adv. 4, eaap7885 (2018).

    Google Scholar 

  59. You, J., Liu, B., Ying, Z., Pande, V. & Leskovec, J. Graph convolutional policy network for goal-directed molecular graph generation. Adv. Neural Inf. Process. Syst. 31, 6410–6421 (2018).

    Google Scholar 

  60. Fu, T., Gao, W., Coley, C. & Sun, J. Reinforced genetic algorithm for structure-based drug design. Adv. Neural Inf. Process. Syst. 35, 12325–12338 (2022).

  61. Fu, T., Xiao, C., Li, X., Glass, L. M. & Sun, J. MIMOSA: multi-constraint molecule sampling for molecule optimization. In Proc. 35th AAAI Conference on Artificial Intelligence 125–133 (AAAI, 2021).

  62. Fu, T. et al. Differentiable scaffolding tree for molecular optimization. In Proc. 10th International Conference on Learning Representations (OpenReview, 2022).

  63. Jin, W., Barzilay, R. & Jaakkola, T. Junction tree variational autoencoder for molecular graph generation. In Proc. 35th International Conference on Machine Learning, Proc. Machine Learning Research Vol. 80 (eds Dy, J. & Krause, A.) 2323–2332 (PMLR, 2018).

  64. Griffiths, R.-R. & Hernández-Lobato, J. M. Constrained Bayesian optimization for automatic chemical design using variational autoencoders. Chem. Sci. 11, 577–586 (2020).

    Google Scholar 

  65. Gao, W., Fu, T., Sun, J. & Coley, C. Sample efficiency matters: a benchmark for practical molecular optimization. Adv. Neural Inf. Process. Syst. 35, 21342–21357 (2022). One of the benchmarks for molecule optimization evaluating more than 25 methods on 20 commonly used oracle functions.

  66. Du, Y. et al. ChemSpacE: interpretable and interactive chemical space exploration. In Transactions on Machine Learning Research (OpenReview, 2023).

  67. Schrödinger release 2024-1 (Schrödinger, 2024).

  68. OpenEye: Applications (Cadence Molecular Sciences, 2023); https://www.eyesopen.com/applications

  69. Friesner, R. A. et al. Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J. Med. Chem. 47, 1739–1749 (2004).

    Google Scholar 

  70. OpenEye: OEDocking (Cadence Molecular Sciences, 2023); https://www.eyesopen.com/oedocking

  71. Arús-Pous, J. et al. Randomized SMILES strings improve the quality of molecular generative models. J. Cheminf. 11, 71 (2019).

    Google Scholar 

  72. Brown, N., Fiscato, M., Segler, M. H. & Vaucher, A. C. GuacaMol: benchmarking models for de novo molecular design. J. Chem. Inf. Model. 59, 1096–1108 (2019). One of the first proposed benchmarks for ML-based molecular design and many proposed tasks are still used in newer benchmarks.

    Google Scholar 

  73. Polykovskiy, D. et al. Molecular sets (MOSES): a benchmarking platform for molecular generation models. Front. Pharmacol. 11, 565644 (2020).

    Google Scholar 

  74. Lipinski, C. A., Lombardo, F., Dominy, B. W. & Feeney, P. J. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Delivery Rev. 23, 3–25 (1997).

    Google Scholar 

  75. Bickerton, G. R., Paolini, G. V., Besnard, J., Muresan, S. & Hopkins, A. L. Quantifying the chemical beauty of drugs. Nat. Chem. 4, 90–98 (2012).

    Google Scholar 

  76. Guo, J. et al. DockStream: a docking wrapper to enhance de novo molecular design. J. Cheminf. 13, 1–21 (2021).

    Google Scholar 

  77. Arnott, J. A. & Planey, S. L. The influence of lipophilicity in drug discovery and design. Expert Opin. Drug Discov. 7, 863–875 (2012).

    Google Scholar 

  78. Hopkins, A. L., Keserü, G. M., Leeson, P. D., Rees, D. C. & Reynolds, C. H. The role of ligand efficiency metrics in drug discovery. Nat. Rev. Drug Discov. 13, 105–121 (2014).

    Google Scholar 

  79. Luo, S., Guan, J., Ma, J. & Peng, J. A 3D generative model for structure-based drug design. Adv. Neural Inf. Process. Syst. 34, 6229–6239 (2021).

  80. Harris, C. et al. Benchmarking generated poses: how rational is structure-based drug design with generative models? Preprint at https://arxiv.org/abs/2308.07413 (2023).

  81. Ertl, P. & Schuffenhauer, A. Estimation of synthetic accessibility score of drug-like molecules based on molecular complexity and fragment contributions. J. Cheminf. 1, 8 (2009).

    Google Scholar 

  82. Fukunishi, Y., Kurosawa, T., Mikami, Y. & Nakamura, H. Prediction of synthetic accessibility based on commercially available compound databases. J. Chem. Inf. Model. 54, 3259–3267 (2014).

    Google Scholar 

  83. Voršilák, M., Kolář, M., Čmelo, I. & Svozil, D. SYBA: Bayesian estimation of synthetic accessibility of organic compounds. J. Cheminf. 12, 35 (2020).

    Google Scholar 

  84. Coley, C. W., Rogers, L., Green, W. H. & Jensen, K. F. SCScore: synthetic complexity learned from a reaction corpus. J. Chem. Inf. Model. 58, 252–261 (2018).

    Google Scholar 

  85. Coley, C. W., Rogers, L., Green, W. H. & Jensen, K. F. Computer-assisted retrosynthesis based on molecular similarity. ACS Cent. Sci. 3, 1237–1245 (2017).

    Google Scholar 

  86. Schwaller, P. et al. Molecular transformer: a model for uncertainty-calibrated chemical reaction prediction. ACS Cent. Sci. 5, 1572–1583 (2019).

    Google Scholar 

  87. Schwaller, P. et al. Predicting retrosynthetic pathways using transformer-based models and a hyper-graph exploration strategy. Chem. Sci. 11, 3316–3325 (2020).

    Google Scholar 

  88. Segler, M. H., Preuss, M. & Waller, M. P. Planning chemical syntheses with deep neural networks and symbolic AI. Nature 555, 604–610 (2018). Seminal work applying deep learning for retrosynthesis, which can be used to filter ML-generated molecules for synthesizability.

    Google Scholar 

  89. Genheden, S. et al. AiZynthFinder: a fast, robust and flexible open-source software for retrosynthetic planning. J. Cheminf. 12, 70 (2020).

    Google Scholar 

  90. Thakkar, A., Chadimová, V., Bjerrum, E. J., Engkvist, O. & Reymond, J.-L. Retrosynthetic accessibility score (RAscore)—rapid machine learned synthesizability classification from AI driven retrosynthetic planning. Chem. Sci. 12, 3339–3349 (2021).

    Google Scholar 

  91. Liu, C.-H. et al. RetroGNN: fast estimation of synthesizability for virtual screening and de novo design by learning from slow retrosynthesis software. J. Chem. Inf. Model. 62, 2293–2300 (2022).

    Google Scholar 

  92. Bradshaw, J., Paige, B., Kusner, M. J., Segler, M. & Hernández-Lobato, J. M. A model to search for synthesizable molecules. Adv. Neural Inf. Process. Syst. 32, 79377949 (2019).

    Google Scholar 

  93. Bradshaw, J., Paige, B., Kusner, M. J., Segler, M. H. S. & Hernández-Lobato, J. M. Barking up the right tree: an approach to search over molecule synthesis dags. Adv. Neural Inf. Process. Syst. 33, (2020).

  94. Horwood, J. & Noutahi, E. Molecular design in synthetically accessible chemical space via deep reinforcement learning. ACS Omega 5, 32984–32994 (2020).

    Google Scholar 

  95. Gottipati, S. K. et al. Learning to navigate the synthetically accessible chemical space using reinforcement learning. In Proc. 37th International Conference on Machine Learning (eds Daumé H. & Singh, A.) 3668–3679 (PMLR, 2020).

  96. Gao, W., Mercado, R. & Coley, C. W. Amortized tree generation for bottom-up synthesis planning and synthesizable molecular design. In Proc. 10th International Conference on Learning Representations (OpenReview, 2022).

  97. Swanson, K. et al. Generative AI for designing and validating easily synthesizable and structurally novel antibiotics. Nat. Mach. Intell. 6, 338–353 (2024).

    Google Scholar 

  98. Fialková, V. et al. LibINVENT: reaction-based generative scaffold decoration for in silico library design. J. Chem. Inf. Model. 62, 2046–2063 (2021).

    Google Scholar 

  99. Hartenfeller, M. et al. DOGS: reaction-driven de novo design of bioactive compounds. PLoS Comput. Biol. 8, e1002380 (2012).

    Google Scholar 

  100. Ghiandoni, G. M. et al. RENATE: a pseudo-retrosynthetic tool for synthetically accessible de novo design. Mol. Inform. 41, 2100207 (2022).

    Google Scholar 

  101. Flam-Shepherd, D., Zhu, K. & Aspuru-Guzik, A. Language models can learn complex molecular distributions. Nat. Commun. 13, 3293 (2022).

    Google Scholar 

  102. Ballarotto, M. et al. De novo design of Nurr1 agonists via fragment-augmented generative deep learning in low-data regime. J. Med. Chem. 66, 8170–8177 (2023).

    Google Scholar 

  103. Gaulton, A. et al. ChEMBL: a large-scale bioactivity database for drug discovery. Nucleic Acids Res. 40, D1100–D1107 (2012).

    Google Scholar 

  104. Ivanenkov, Y. A. et al. Chemistry42: an AI-driven platform for molecular design and optimization. J. Chem. Info. Model. 63, 695–701 (2023).

    Google Scholar 

  105. Zhu, W. et al. Discovery of novel and selective SIK2 inhibitors by the application of AlphaFold structures and generative models. Bioorg. Med. Chem. 91, 117414 (2023).

    Google Scholar 

  106. Grisoni, F. et al. Combining generative artificial intelligence and on-chip synthesis for de novo drug design. Sci. Adv. 7, eabg3338 (2021).

    Google Scholar 

  107. Li, Y. et al. Generative deep learning enables the discovery of a potent and selective RIPK1 inhibitor. Nat. Commun. 13, 6891 (2022). A study showing the complementarity of virtual screening and generative molecular design with experimental validation.

    Google Scholar 

  108. Yoshimori, A. et al. Design and synthesis of DDR1 inhibitors with a desired pharmacophore using deep generative models. ChemMedChem 16, 955–958 (2021).

    Google Scholar 

  109. Blaschke, T. et al. REINVENT 2.0: an AI tool for de novo drug design. J. Chem. Inf. Model. 60, 5918–5922 (2020).

    Google Scholar 

  110. Merk, D., Grisoni, F., Friedrich, L., Gelzinyte, E. & Schneider, G. Scaffold hopping from synthetic RXR modulators by virtual screening and de novo design. MedChemComm 9, 1289–1292 (2018).

    Google Scholar 

  111. Merk, D., Grisoni, F., Friedrich, L., Gelzinyte, E. & Schneider, G. Computer-assisted discovery of retinoid X receptor modulating natural products and isofunctional mimetics. J. Med. Chem. 61, 5442–5447 (2018).

    Google Scholar 

  112. Gillet, V. et al. Synthetically accessible de novo design using reaction vectors: application to PARP1 inhibitors. Mol. Inform. 43, e202300183 (2024).

    Google Scholar 

  113. Pun, F. W., Ozerov, I. V. & Zhavoronkov, A. AI-powered therapeutic target discovery. Trends Pharmacol. Sci. 44, 561–572 (2023).

    Google Scholar 

  114. First generative AI drug begins phase II trials with patients. Insilico Medicine https://insilico.com/blog/first_phase2 (2023).

  115. Ren, F. et al. A small-molecule TNIK inhibitor targets fibrosis in preclinical and clinical models. Nat. Biotechnol. (2024).

  116. Guo, J. et al. Improving de novo molecular design with curriculum learning. Nat. Mach. Intell. 4, 555–563 (2022).

    Google Scholar 

  117. Guo, J. & Schwaller, P. Augmented memory: capitalizing on experience replay to accelerate de novo molecular design. Preprint at https://doi.org/10.48550/arXiv.2305.16160 (2024).

  118. Guo, J. & Schwaller, P. Beam enumeration: probabilistic explainability for sample efficient self-conditioned molecular design. In Proc. 12th International Conference on Learning Representations (OpenReview, 2024).

  119. Dodds, M. et al. Sample efficient reinforcement learning with active learning for molecular design. Chem. Sci. 15, 4146–4160 (2024).

    Google Scholar 

  120. Buttenschoen, M., Morris, G. M. & Deane, C. M. Posebusters: AI-based docking methods fail to generate physically valid poses or generalise to novel sequences. Chem. Sci. 15, 3130–3139 (2024).

    Google Scholar 

  121. Watson, J. L. et al. De novo design of protein structure and function with RFdiffusion. Nature 620, 1089–1100 (2023).

    Google Scholar 

  122. Du, Y., Guo, X., Wang, Y., Shehu, A. & Zhao, L. Small molecule generation via disentangled representation learning. Bioinformatics 38, 3200–3208 (2022).

    Google Scholar 

  123. Jin, W., Barzilay, R. & Jaakkola, T. Multi-objective molecule generation using interpretable substructures. In Proc. 37th International Conference on Machine Learning 4849–4859 (PMLR, 2020).

  124. Hoffman, S. C., Chenthamarakshan, V., Wadhawan, K., Chen, P.-Y. & Das, P. Optimizing molecules using efficient queries from property evaluations. Nat. Mach. Intell. 4, 21–31 (2021).

    Google Scholar 

  125. Madhawa, K., Ishiguro, K., Nakago, K. & Abe, M. GraphNVP: an invertible flow model for generating molecular graphs. Preprint at https://arxiv.org/abs/1905.11600 (2019).

  126. Kadurin, A. et al. The cornucopia of meaningful leads: applying deep adversarial autoencoders for new molecule development in oncology. Oncotarget 8, 10883 (2017).

    Google Scholar 

  127. Imrie, F., Bradley, A. R., van der Schaar, M. & Deane, C. M. Deep generative models for 3D linker design. J. Chem. Inf. Model. 60, 1983–1995 (2020).

    Google Scholar 

  128. Liu, M., Yan, K., Oztekin, B. & Ji, S. GraphEBM: molecular graph generation with energy-based models. Preprint at https://arxiv.org/abs/2102.00546 (2021).

  129. Vignac, C. et al. DiGress: discrete denoising diffusion for graph generation. In Proc. 11th International Conference on Learning Representations (OpenReview, 2023).

  130. Nigam, A., Friederich, P., Krenn, M. & Aspuru-Guzik, A. Augmenting genetic algorithms with deep neural networks for exploring the chemical space. In Proc. 8th International Conference on Learning Representations (OpenReview, 2020).

  131. Spiegel, J. O. & Durrant, J. D. AutoGrow4: an open-source genetic algorithm for de novo drug design and lead optimization. J. Cheminf. 12, 25 (2020). One of the representative works that leverage genetic algorithms for molecular design.

    Google Scholar 

  132. Simm, G. & Hernandez-Lobato, J. M. A generative model for molecular distance geometry. In Proc. 37th International Conference on Machine Learning, Proc. Machine Learning Research Vol. 119 (eds Daumé H. & Singh, A.) 8949–8958 (PMLR, 2020).

  133. Ganea, O. et al. GeoMol: torsional geometric generation of molecular 3D conformer ensembles. Adv. Neural Inf. Process. Syst. 34, 13757–13769 (2021).

  134. Klein, L., Krämer, A. & Noe, F. Equivariant flow matching. Adv. Neural Inf. Process. Syst. 36, 59886–59910 (2023).

  135. Stärk, H., Ganea, O., Pattanaik, L., Barzilay, D. & Jaakkola, T. EquiBind: geometric deep learning for drug binding structure prediction. In Proc. 39th International Conference on Machine Learning, Proc. Machine Learning Research Vol. 162 (eds Chaudhuri, K. et al.) 20503–20521 (PMLR, 2022).

  136. Jing, B., Corso, G., Chang, J., Barzilay, R. & Jaakkola, T. Torsional diffusion for molecular conformer generation. Adv. Neural Inf. Process. Syst. 35, 24240–24253 (2022).

  137. Corso, G., Stärk, H., Jing, B., Barzilay, R. & Jaakkola, T. S. DiffDock: diffusion steps, twists, and turns for molecular docking. In Proc. 11th International Conference on Learning Representations (OpenReview, 2023).

  138. Ragoza, M., Masuda, T. & Koes, D. R. Generating 3D molecules conditional on receptor binding sites with deep generative models. Chem. Sci. 13, 2701–2713 (2022).

    Google Scholar 

  139. Drotár, P., Jamasb, A. R., Day, B., Cangea, C. & Liò, P. Structure-aware generation of drug-like molecules. Preprint at https://arxiv.org/abs/2111.04107 (2021).

  140. Joshi, R. P. et al. 3D-scaffold: a deep learning framework to generate 3D coordinates of drug-like molecules with desired scaffolds. J. Phys. Chem. B 125, 12166–12176 (2021).

    Google Scholar 

  141. Liu, M., Luo, Y., Uchino, K., Maruhashi, K. & Ji, S. Generating 3D molecules for target protein binding. In Proc. 39th International Conference on Machine Learning, Proc. Machine Learning Research, Vol. 162 (eds Chaudhuri, K. et al) 13912–13924 (PMLR, 2022).

  142. Garcia Satorras, V., Hoogeboom, E., Fuchs, F., Posner, I. & Welling, M. E(n) equivariant normalizing flows. Adv. Neural Inf. Process. Syst. 34, 4181–4192 (2021).

  143. Graves, A. et al. Hybrid computing using a neural network with dynamic external memory. Nature 538, 471–476 (2016).

    Google Scholar 

  144. Makhzani, A., Shlens, J., Jaitly, N., Goodfellow, I. & Frey, B. Adversarial autoencoders. Preprint at https://arxiv.org/abs/1511.05644 (2015).

  145. Philippidis, A. Insilico joins scramble to treat solid tumors by targeting KIF18A; https://www.genengnews.com/topics/artificial-intelligence/insilico-joins-scramble-to-treat-solid-tumors-by-targeting-kif18a/ (2024).

  146. Merk, D., Friedrich, L., Grisoni, F. & Schneider, G. De novo design of bioactive small molecules by artificial intelligence. Mol. Inf. 37, 1700153 (2018).

    Google Scholar 

  147. Merk, D., Grisoni, F., Friedrich, L. & Schneider, G. Tuning artificial intelligence on the de novo design of natural-product-inspired retinoid X receptor modulators. Commun. Chem. 1, 68 (2018).

    Google Scholar 

  148. Yu, Y. et al. A novel scalarized scaffold hopping algorithm with graph-based variational autoencoder for discovery of JAK1 inhibitors. ACS Omega 6, 22945–22954 (2021).

    Google Scholar 

  149. Moret, M., Helmstädter, M., Grisoni, F., Schneider, G. & Merk, D. Beam search for automated design and scoring of novel ROR ligands with machine intelligence. Angew. Chem. Int. Ed. 60, 19477–19482 (2021).

    Google Scholar 

  150. Jang, S. H. et al. PCW-A1001, AI-assisted de novo design approach to design a selective inhibitor for FLT-3 (D835Y) in acute myeloid leukemia. Front. Mol. Biosci. 9, 1072028 (2022).

    Google Scholar 

  151. Eguida, M., Schmitt-Valencia, C., Hibert, M., Villa, P. & Rognan, D. Target-focused library design by pocket-applied computer vision and fragment deep generative linking. J. Med. Chem. 65, 13771–13783 (2022).

    Google Scholar 

  152. Chen, N. et al. Recurrent neural network (RNN) model accelerates the development of antibacterial metronidazole derivatives. RSC Adv. 12, 22893–22901 (2022).

    Google Scholar 

  153. Tan, X. et al. Discovery of pyrazolo [3,4-d] pyridazinone derivatives as selective DDR1 inhibitors via deep learning based design, synthesis, and biological evaluation. J. Med. Chem. 65, 103–119 (2021).

    Google Scholar 

  154. Hua, Y. et al. Effective reaction-based de novo strategy for kinase targets: a case study on MERTK inhibitors. J. Chem. Inf. Model. 62, 1654–1668 (2022).

    Google Scholar 

  155. Moret, M. et al. Leveraging molecular structure and bioactivity with chemical language models for de novo drug design. Nat. Commun. 14, 114 (2023).

    Google Scholar 

  156. Song, S. et al. Application of deep generative model for design of pyrrolo [2,3-d] pyrimidine derivatives as new selective tank binding kinase 1 (TBK1) inhibitors. Eur. J. Med. Chem. 247, 115034 (2023).

    Google Scholar 

  157. Yu, Y. et al. Accelerated discovery of macrocyclic CDK2 inhibitor QR-6401 by generative models and structure-based drug design. ACS Med. Chem. Lett. 14, 297–304 (2023).

    Google Scholar 

  158. Atz, K. et al. Prospective de novo drug design with deep interactome learning. Nat. Commun. 15, 3408 (2024).

    Google Scholar 

  159. Putin, E. et al. Adversarial threshold neural computer for molecular de novo design. Mol. Pharmaceutics 15, 4386–4397 (2018).

    Google Scholar 

  160. Polykovskiy, D. et al. Entangled conditional adversarial autoencoder for de novo drug discovery. Mol. Pharm. 15, 4398–4405 (2018).

    Google Scholar 

  161. Korshunova, M. et al. Generative and reinforcement learning approaches for the automated de novo design of bioactive compounds. Commun. Chem. 5, 129 (2022).

    Google Scholar 

  162. Li, Y. et al. Discovery of potent, selective, and orally bioavailable small-molecule inhibitors of CDK8 for the treatment of cancer. J. Med. Chem. 66, 5439–5452 (2023).

    Google Scholar 

  163. Salas-Estrada, L. et al. De novo design of κ-opioid receptor antagonists using a generative deep-learning framework. J. Chem. Inf. Model. 63, 5056–5065 (2023).

    Google Scholar 

  164. Xu, J. et al. Discovery of novel and potent prolyl hydroxylase domain-containing protein (PHD) inhibitors for the treatment of anemia. J. Med. Chem. 67, 1393–1405 (2024).

    Google Scholar 

  165. Bo, W. et al. Local scaffold diversity-contributed generator for discovering potential NLRP3 inhibitors. J. Chem. Inf. Model. 64, 737–748 (2024).

    Google Scholar 

  166. Xia, Y. et al. Target-aware molecule generation for drug design using a chemical language model. Preprint at bioRxiv https://doi.org/10.1101/2024.01.08.574635 (2024).

  167. Vakili, M. G. et al. Quantum computing-enhanced algorithm unveils novel inhibitors for KRAS. Preprint at https://arxiv.org/abs/2402.08210 (2024).

  168. Hassen, A. K. et al. Generate what you can make: achieving in-house synthesizability with readily available resources in de novo drug design. Preprint at chemRxiv https://doi.org/10.26434/chemrxiv-2024-wtjt6 (2024).

  169. Wang, Y. et al. Discovery of 3-hydroxymethyl-azetidine derivatives as potent polymerase theta inhibitors. Bioorg. Med. Chem. 103, 117662 (2024).

    Google Scholar 

  170. Zhao, Y. et al. Accelerating factor Xa inhibitor discovery with a de novo drug design pipeline. Chin. J. Chem. Eng. (2024).

  171. Jiang, Y. et al. Pocketflow is a data-and-knowledge-driven structure-based molecular generative model. Nat. Mach. Intell. 6, 326–337 (2024).

    Google Scholar 

  172. Zhang, J. et al. ISM9682A, a novel and potent KIF18A inhibitor, shows robust antitumor effects against chromosomally unstable cancers. Cancer Res. 84, 5727–5727 (2024).

    Google Scholar 

  173. Huang, L. et al. A dual diffusion model enables 3D molecule generation and lead optimization based on target pockets. Nat. Commun. 15, 2657 (2024).

    Google Scholar 

Download references

Acknowledgements

J.G. and P.S. acknowledge support from the NCCR Catalysis (grant number 180544), a National Centre of Competence in Research funded by the Swiss National Science Foundation. A.R.J. is funded by a Biotechnology and Biological Sciences Research Council (BBSRC) DTP studentship (BB/M011194/1). Y.W. acknowledges the support of Cornell Presidential Life Science Fellowship. We are grateful to K. Atz and A. Mueller for helpful feedback and discussion.

Author information

Author notes

  1. Arian R. Jamasb

    Present address: Prescient Design, Genentech, Basel, Switzerland

  2. These authors contributed equally: Yuanqi Du, Arian R. Jamasb, Jeff Guo.

Authors and Affiliations

  1. Department of Computer Science, Cornell University, Ithaca, NY, USA

    Yuanqi Du & Yingheng Wang

  2. Department of Biochemistry, University of Cambridge, Cambridge, UK

    Arian R. Jamasb

  3. Department of Computer Science and Technology, University of Cambridge, Cambridge, UK

    Arian R. Jamasb, Charles Harris & Pietro Liò

  4. Laboratory of Artificial Chemical Intelligence (LIAC), Institut des Sciences et Ingénierie Chimiques, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland

    Jeff Guo & Philippe Schwaller

  5. National Centre of Competence in Research (NCCR) Catalysis, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland

    Jeff Guo & Philippe Schwaller

  6. Rensselaer Polytechnic Institute, Troy, NY, USA

    Tianfan Fu

  7. Microsoft Quantum, Redmond, WA, USA

    Chenru Duan

  8. Heart and Lung Research Institute, University of Cambridge, Cambridge, UK

    Tom L. Blundell

Authors
  1. Yuanqi Du
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Arian R. Jamasb
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jeff Guo
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Tianfan Fu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Charles Harris
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yingheng Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Chenru Duan
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Pietro Liò
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Philippe Schwaller
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Tom L. Blundell
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Y.D., A.R.J. and J.G. led this work under the supervision of P.S. and T.L.B. and contributed equally. T.F. and C.H. also contributed equally. All authors contributed ideas and discussions to writing, reviewing and editing of the paper before submission.

Corresponding authors

Correspondence to Philippe Schwaller or Tom L. Blundell.

Ethics declarations

Competing interests

A.R.J. declares a potential financial conflict of interest due to his role as a machine learning scientist at Prescient Design, Genentech. The other authors declare no competing interests.

Peer review

Peer review information

Nature Machine Intelligence thanks J. B. Brown and Ola Engkvist for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary text and references.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Du, Y., Jamasb, A.R., Guo, J. et al. Machine learning-aided generative molecular design. Nat Mach Intell 6, 589–604 (2024). https://doi.org/10.1038/s42256-024-00843-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s42256-024-00843-5

  • Springer Nature Limited

This article is cited by

Search

Navigation