{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T23:08:47Z","timestamp":1789600127490,"version":"build-2803163510"},"reference-count":65,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,8]],"date-time":"2026-06-08T00:00:00Z","timestamp":1780876800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T00:00:00Z","timestamp":1789603200000},"content-version":"vor","delay-in-days":101,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Fujian Sunshine Charity Foundation of China"},{"name":"National Key R&D Program of China"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Commun Chem"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Designing drug-like molecules that satisfy both properties requirements and structural constraints remains challenging. Current generative approaches typically reduce properties to numerical constraints and treat molecular structures as deterministic functions of properties, failing to capture complex, nonlinear structure-property relationships and limiting generative diversity and controllability. Therefore, we propose BiSP-GP, a bidirectional structure-property generative pretraining framework that unifies molecular generation and property prediction as a single sequence modeling task. BiSP-GP serializes continuous properties into semantic token sequences for joint modeling with molecular structures in a shared sequence space. Molecular generation and property prediction are cast as autoregressive sequence modeling tasks, with a cross-modal decoder supporting bidirectional mapping. The framework also incorporates scaffolds as conditional inputs to guide structure generation. Experiments demonstrate that BiSP-GP achieves strong performance gconditional molecular generation, property prediction, and downstream tasks. A case study on PAK1 further validates the model\u2019s generative ability and shows improved binding capacity in molecular docking evaluation.<\/jats:p>","DOI":"10.1038\/s42004-026-02093-5","type":"journal-article","created":{"date-parts":[[2026,6,8]],"date-time":"2026-06-08T18:06:16Z","timestamp":1780941976000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Generative pretraining for drug molecule design with bidirectional structure-property optimization"],"prefix":"10.1038","volume":"9","author":[{"given":"Yingying","family":"Song","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4136-6151","authenticated-orcid":false,"given":"Song","family":"He","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3490-5812","authenticated-orcid":false,"given":"Xiaochen","family":"Bo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7227-3943","authenticated-orcid":false,"given":"Zhongnan","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,8]]},"reference":[{"key":"2093_CR1","doi-asserted-by":"publisher","first-page":"1241","DOI":"10.1016\/j.drudis.2018.01.039","volume":"23","author":"H Chen","year":"2018","unstructured":"Chen, H. et al. The rise of deep learning in drug discovery. Drug Discov. Today.\u00a023, 1241\u20131250 (2018).","journal-title":"Drug Discov. Today"},{"key":"2093_CR2","doi-asserted-by":"publisher","first-page":"1074","DOI":"10.1038\/s42003-024-06746-w","volume":"7","author":"Y Yang","year":"2024","unstructured":"Yang, Y. et al. Enabling target-aware molecule generation to follow multi objectives with Pareto MCTS. Commun. Biol. 7, 1074 (2024).","journal-title":"Commun. Biol."},{"key":"2093_CR3","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-025-57582-3","volume":"16","author":"T Yoshizawa","year":"2025","unstructured":"Yoshizawa, T. et al. A data-driven generative strategy to avoid reward hacking in multi-objective molecular design. Nat. Commun. 16, 2409 (2025).","journal-title":"Nat. Commun."},{"key":"2093_CR4","doi-asserted-by":"publisher","first-page":"238","DOI":"10.1038\/s42004-025-01635-7","volume":"8","author":"I Filella-Merce","year":"2025","unstructured":"Filella-Merce, I. et al. Optimizing drug design by merging generative AI with a physics-based active learning framework. Commun. Chem. 8, 238 (2025).","journal-title":"Commun. Chem."},{"key":"2093_CR5","unstructured":"Luo, Y., Yan, K. & Ji, S. Graphdf: a discrete flow model for molecular graph generation. In International conference on machine learning. 7192\u20137203 (PMLR, 2021)."},{"key":"2093_CR6","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-47120-y","volume":"15","author":"BP Munson","year":"2024","unstructured":"Munson, B. P. et al. De novo generation of multi-target compounds using deep generative chemistry. Nat. Commun. 15, 3636 (2024).","journal-title":"Nat. Commun."},{"key":"2093_CR7","doi-asserted-by":"publisher","first-page":"9478","DOI":"10.1021\/acs.jmedchem.2c00732","volume":"65","author":"M Wang","year":"2022","unstructured":"Wang, M. et al. Relation: A deep generative model for structure-based de novo drug design. J. Med. Chem. 65, 9478\u20139492 (2022).","journal-title":"J. Med. Chem."},{"key":"2093_CR8","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-28526-y","volume":"13","author":"NW Gebauer","year":"2022","unstructured":"Gebauer, N. W. et al. Inverse design of 3d molecular structures with conditional generative neural networks. Nat. Commun. 13, 973 (2022).","journal-title":"Nat. Commun."},{"key":"2093_CR9","doi-asserted-by":"publisher","first-page":"120","DOI":"10.1021\/acscentsci.7b00512","volume":"4","author":"MH Segler","year":"2018","unstructured":"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\u2013131 (2018).","journal-title":"ACS Cent. Sci."},{"key":"2093_CR10","doi-asserted-by":"publisher","first-page":"4059","DOI":"10.1021\/acs.jcim.4c00504","volume":"64","author":"R Gou","year":"2024","unstructured":"Gou, R. et al. CNSMolGen: a bidirectional recurrent neural network-based generative model for de novo central nervous system drug design. J. Chem. Inf. Modeling. 64, 4059\u20134070 (2024).","journal-title":"J. Chem. Inf. Modeling"},{"key":"2093_CR11","doi-asserted-by":"publisher","first-page":"5844","DOI":"10.1021\/acs.jcim.4c00531","volume":"64","author":"Y Matsukiyo","year":"2024","unstructured":"Matsukiyo, Y., Tengeiji, A., Li, C. & Yamanishi, Y. Transcriptionally conditional recurrent neural network for de novo drug design. J. Chem. Inf. Modeling. 64, 5844\u20135852 (2024).","journal-title":"J. Chem. Inf. Modeling"},{"key":"2093_CR12","doi-asserted-by":"publisher","first-page":"btac814","DOI":"10.1093\/bioinformatics\/btac814","volume":"39","author":"Z Liao","year":"2023","unstructured":"Liao, Z., Xie, L., Mamitsuka, H. & Zhu, S. Sc2Mol: a scaffold-based two-step molecule generator with variational autoencoder and transformer. Bioinformatics. 39, btac814 (2023).","journal-title":"Bioinformatics"},{"key":"2093_CR13","doi-asserted-by":"publisher","unstructured":"Yoshikai, Y., Mizuno, T., Nemoto, S. & Kusuhara, H. A novel molecule generative model of VAE combined with Transformer for unseen structure generation. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2402.11950 (2024).","DOI":"10.48550\/arXiv.2402.11950"},{"key":"2093_CR14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-019-0404-1","volume":"12","author":"\u0141 Maziarka","year":"2020","unstructured":"Maziarka, \u0141 et al. Mol-CycleGAN: a generative model for molecular optimization. J. Cheminform. 12, 1\u201318 (2020).","journal-title":"J. Cheminform."},{"key":"2093_CR15","doi-asserted-by":"publisher","first-page":"5316","DOI":"10.1039\/D1CC07035E","volume":"58","author":"B Sridharan","year":"2022","unstructured":"Sridharan, B., Goel, M. & Priyakumar, U. D. Modern machine learning for tackling inverse problems in chemistry: molecular design to realization. Chem. Commun. 58, 5316\u20135331 (2022).","journal-title":"Chem. Commun."},{"key":"2093_CR16","doi-asserted-by":"publisher","first-page":"e1012229","DOI":"10.1371\/journal.pcbi.1012229","volume":"20","author":"C Ai","year":"2024","unstructured":"Ai, C. et al. MTMol-GPT: De novo multi-target molecular generation with transformer-based generative adversarial imitation learning. PLoS Comput. Biol. 20, e1012229 (2024).","journal-title":"PLoS Comput. Biol."},{"key":"2093_CR17","doi-asserted-by":"publisher","first-page":"1153","DOI":"10.1039\/C9SC04503A","volume":"11","author":"J Lim","year":"2020","unstructured":"Lim, J. et al. Scaffold-based molecular design with a graph generative model. Chem. Sci. 11, 1153\u20131164 (2020).","journal-title":"Chem. Sci."},{"key":"2093_CR18","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-025-59439-1","volume":"16","author":"F Therrien","year":"2025","unstructured":"Therrien, F., Sargent, E. H. & Voznyy, O. Using GNN property predictors as molecule generators. Nat. Commun. 16, 4301 (2025).","journal-title":"Nat. Commun."},{"key":"2093_CR19","unstructured":"Jin, W., Barzilay, R. & Jaakkola, T. Multi-objective molecule generation using interpretable substructures. In International conference on machine learning. 4849\u20134859 (PMLR, 2020)."},{"key":"2093_CR20","doi-asserted-by":"crossref","unstructured":"Li, C., Tang, H., Zhu, Y. & Yamanishi, Y. A Reinforcement Learning-Driven Transformer GAN for Molecular Generation. Machine Intelligence Research. 1\u201322 (2026)","DOI":"10.1007\/s11633-025-1589-6"},{"key":"2093_CR21","doi-asserted-by":"publisher","first-page":"1700123","DOI":"10.1002\/minf.201700123","volume":"37","author":"T Blaschke","year":"2018","unstructured":"Blaschke, T. et al. Application of generative autoencoder in de novo molecular design. Mol. Inform. 37, 1700123 (2018).","journal-title":"Mol. Inform."},{"key":"2093_CR22","doi-asserted-by":"publisher","first-page":"48906","DOI":"10.52202\/075280-2124","volume":"36","author":"S Lee","year":"2023","unstructured":"Lee, S. et al. Advancing bayesian optimization via learning correlated latent space. Adv. Neural Inf. Process. Syst. 36, 48906\u201348917 (2023).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"2093_CR23","doi-asserted-by":"publisher","first-page":"254","DOI":"10.1038\/s42256-020-0174-5","volume":"2","author":"P-C Kotsias","year":"2020","unstructured":"Kotsias, P.-C. et al. Direct steering of de novo molecular generation with descriptor conditional recurrent neural networks. Nat. Mach. Intell. 2, 254\u2013265 (2020).","journal-title":"Nat. Mach. Intell."},{"key":"2093_CR24","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-018-0286-7","volume":"10","author":"J Lim","year":"2018","unstructured":"Lim, J., Ryu, S., Kim, J. W. & Kim, W. Y. Molecular generative model based on conditional variational autoencoder for de novo molecular design. J. Cheminform. 10, 1\u20139 (2018).","journal-title":"J. Cheminform."},{"key":"2093_CR25","doi-asserted-by":"publisher","first-page":"2943","DOI":"10.1021\/acs.jcim.2c00487","volume":"62","author":"M Lee","year":"2022","unstructured":"Lee, M. & Min, K. Mgcvae: multi-objective inverse design via molecular graph conditional variational autoencoder. J. Chem. Inf. Modeling.\u00a062, 2943\u20132950 (2022).","journal-title":"J. Chem. Inf. Modeling"},{"key":"2093_CR26","unstructured":"Yu, H. & Yu, H. TensorVAE: a simple and efficient generative model for conditional molecular conformation generation. Trans. Mach. Learn. Res.2024 (2024)."},{"key":"2093_CR27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-018-0287-6","volume":"10","author":"Y Li","year":"2018","unstructured":"Li, Y., Zhang, L. & Liu, Z. Multi-objective de novo drug design with conditional graph generative model. J. Cheminform. 10, 1\u201324 (2018).","journal-title":"J. Cheminform."},{"key":"2093_CR28","doi-asserted-by":"crossref","unstructured":"Huang, H., Sun, L., Du, B. & Lv, W. Conditional diffusion based on discrete graph structures for molecular graph generation. In Proc. of the AAAI Conference on Artificial Intelligence. 4302-4311 (AAAI Press, 2023).","DOI":"10.1609\/aaai.v37i4.25549"},{"key":"2093_CR29","doi-asserted-by":"publisher","first-page":"2064","DOI":"10.1021\/acs.jcim.1c00600","volume":"62","author":"V Bagal","year":"2021","unstructured":"Bagal, V., Aggarwal, R., Vinod, P. & Priyakumar, U. D. MolGPT: molecular generation using a transformer-decoder model. J. Chem. Inf. Modeling. 62, 2064\u20132076 (2021).","journal-title":"J. Chem. Inf. Modeling"},{"key":"2093_CR30","doi-asserted-by":"publisher","first-page":"914","DOI":"10.1038\/s42256-021-00403-1","volume":"3","author":"J Wang","year":"2021","unstructured":"Wang, J. et al. Multi-constraint molecular generation based on conditional transformer, knowledge distillation and reinforcement learning. Nat. Mach. Intell. 3, 914\u2013922 (2021).","journal-title":"Nat. Mach. Intell."},{"key":"2093_CR31","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-23415-2","volume":"12","author":"O Mahmood","year":"2021","unstructured":"Mahmood, O., Mansimov, E., Bonneau, R. & Cho, K. Masked graph modeling for molecule generation. Nat. Commun. 12, 3156 (2021).","journal-title":"Nat. Commun."},{"key":"2093_CR32","doi-asserted-by":"publisher","first-page":"015022","DOI":"10.1088\/2632-2153\/ac3ffb","volume":"3","author":"R Irwin","year":"2022","unstructured":"Irwin, R., Dimitriadis, S., He, J. & Bjerrum, E. J. Chemformer: a pre-trained transformer for computational chemistry. Mach. Learn. Sci. Technol. 3, 015022 (2022).","journal-title":"Mach. Learn. Sci. Technol."},{"key":"2093_CR33","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-46440-3","volume":"15","author":"J Chang","year":"2024","unstructured":"Chang, J. & Ye, J. C. Bidirectional generation of structure and properties through a single molecular foundation model. Nat. Commun. 15, 2323 (2024).","journal-title":"Nat. Commun."},{"key":"2093_CR34","doi-asserted-by":"crossref","unstructured":"Nguyen, C. Q., Kreatsoulas, C. & Branson, K. M. Meta-learning GNN initializations for low-resource molecular property prediction. In 4th Lifelong Machine Learning Workshop at ICML. (2020).","DOI":"10.26434\/chemrxiv.11981622.v1"},{"key":"2093_CR35","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-024-00904-2","volume":"16","author":"M Garc\u00eda-Orteg\u00f3n","year":"2024","unstructured":"Garc\u00eda-Orteg\u00f3n, M. et al. Graph neural processes for molecules: an evaluation on docking scores and strategies to improve generalization. J. Cheminform. 16, 115 (2024).","journal-title":"J. Cheminform."},{"key":"2093_CR36","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1021\/acs.chemrestox.5c00033","volume":"38","author":"S Seal","year":"2025","unstructured":"Seal, S. et al. Machine learning for toxicity prediction using chemical structures: pillars for success in the real world. Chem. Res. Toxicol. 38, 759\u2013807 (2025).","journal-title":"Chem. Res. Toxicol."},{"key":"2093_CR37","doi-asserted-by":"publisher","first-page":"1336025","DOI":"10.3389\/fddsv.2024.1336025","volume":"4","author":"L Tonoyan","year":"2024","unstructured":"Tonoyan, L. & Siraki, A. G. Machine learning in toxicological sciences: opportunities for assessing drug toxicity. Front. Drug Discov. 4, 1336025 (2024).","journal-title":"Front. Drug Discov."},{"key":"2093_CR38","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-023-00796-8","volume":"16","author":"S Lee","year":"2024","unstructured":"Lee, S. & Yoo, S. InterDILI: interpretable prediction of drug-induced liver injury through permutation feature importance and attention mechanism. J. Cheminform. 16, 1 (2024).","journal-title":"J. Cheminform."},{"key":"2093_CR39","doi-asserted-by":"publisher","first-page":"1290","DOI":"10.1021\/acs.chemrestox.4c00015","volume":"37","author":"S Seal","year":"2024","unstructured":"Seal, S. et al. Improved detection of drug-induced liver injury by integrating predicted in vivo and in vitro data. Chem. Res. Toxicol. 37, 1290\u20131305 (2024).","journal-title":"Chem. Res. Toxicol."},{"key":"2093_CR40","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-020-18305-y","volume":"11","author":"D Galeano","year":"2020","unstructured":"Galeano, D., Li, S., Gerstein, M. & Paccanaro, A. Predicting the frequencies of drug side effects. Nat. Commun. 11, 4575 (2020).","journal-title":"Nat. Commun."},{"key":"2093_CR41","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-023-37570-1","volume":"14","author":"N Moshkov","year":"2023","unstructured":"Moshkov, N. et al. Predicting compound activity from phenotypic profiles and chemical structures. Nat. Commun. 14, 1967 (2023).","journal-title":"Nat. Commun."},{"key":"2093_CR42","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-019-0397-9","volume":"11","author":"O Prykhodko","year":"2019","unstructured":"Prykhodko, O. et al. A de novo molecular generation method using latent vector based generative adversarial network. J. Cheminform. 11, 74 (2019).","journal-title":"J. Cheminform."},{"key":"2093_CR43","doi-asserted-by":"publisher","first-page":"2684","DOI":"10.1039\/D5DD00122F","volume":"4","author":"J Ross","year":"2025","unstructured":"Ross, J. et al. GP-MoLFormer: a foundation model for molecular generation. Digit. Discov. 4, 2684\u20132696 (2025).","journal-title":"Digit. Discov."},{"key":"2093_CR44","doi-asserted-by":"publisher","first-page":"435","DOI":"10.1021\/cn100008c","volume":"1","author":"TT Wager","year":"2010","unstructured":"Wager, T. T., Hou, X., Verhoest, P. R. & Villalobos, A. Moving beyond rules: the development of a central nervous system multiparameter optimization (CNS MPO) approach to enable alignment of druglike properties. ACS Chem. Neurosci. 1, 435\u2013449 (2010).","journal-title":"ACS Chem. Neurosci."},{"key":"2093_CR45","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbad185","volume":"24","author":"M Yang","year":"2023","unstructured":"Yang, M. et al. CMGN: a conditional molecular generation net to design target-specific molecules with desired properties. Brief. Bioinform. 24, bbad185 (2023).","journal-title":"Brief. Bioinform."},{"key":"2093_CR46","doi-asserted-by":"publisher","first-page":"7253","DOI":"10.1021\/acs.jmedchem.2c01824","volume":"66","author":"FM Cornelissen","year":"2023","unstructured":"Cornelissen, F. M. et al. Explaining blood\u2013brain barrier permeability of small molecules by integrated analysis of different transport mechanisms. J. Med. Chem. 66, 7253\u20137267 (2023).","journal-title":"J. Med. Chem."},{"key":"2093_CR47","doi-asserted-by":"publisher","first-page":"112517","DOI":"10.1016\/j.ejmech.2020.112517","volume":"203","author":"M Zhang","year":"2020","unstructured":"Zhang, M. et al. Design and synthesis of 1H-indazole-3-carboxamide derivatives as potent and selective PAK1 inhibitors with anti-tumour migration and invasion activities. Eur. J. Med. Chem. 203, 112517 (2020).","journal-title":"Eur. J. Med. Chem."},{"key":"2093_CR48","doi-asserted-by":"publisher","first-page":"36372","DOI":"10.1074\/jbc.M505031200","volume":"280","author":"G Zhu","year":"2005","unstructured":"Zhu, G. et al. A single pair of acidic residues in the kinase major groove mediates strong substrate preference for P-2 or P-5 arginine in the AGC, CAMK, and STE kinase families. J. Biol. Chem. 280, 36372\u201336379 (2005).","journal-title":"J. Biol. Chem."},{"key":"2093_CR49","doi-asserted-by":"publisher","first-page":"861","DOI":"10.1002\/jps.21494","volume":"98","author":"R Mannhold","year":"2009","unstructured":"Mannhold, R., Poda, G. I., Ostermann, C. & Tetko, I. V. Calculation of molecular lipophilicity: State-of-the-art and comparison of log P methods on more than 96,000 compounds. J. Pharm. Sci. 98, 861\u2013893 (2009).","journal-title":"J. Pharm. Sci."},{"key":"2093_CR50","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1039\/C7SC02664A","volume":"9","author":"Z Wu","year":"2018","unstructured":"Wu, Z. et al. MoleculeNet: a benchmark for molecular machine learning. Chem. Sci. 9, 513\u2013530 (2018).","journal-title":"Chem. Sci."},{"key":"2093_CR51","doi-asserted-by":"publisher","first-page":"305","DOI":"10.1038\/nature09107","volume":"465","author":"F-J Gamo","year":"2010","unstructured":"Gamo, F.-J. et al. Thousands of chemical starting points for antimalarial lead identification. Nature. 465, 305\u2013310 (2010).","journal-title":"Nature"},{"key":"2093_CR52","doi-asserted-by":"publisher","first-page":"2241","DOI":"10.1021\/jz200866s","volume":"2","author":"J Hachmann","year":"2011","unstructured":"Hachmann, J. et al. The Harvard clean energy project: large-scale computational screening and design of organic photovoltaics on the world community grid.  J. Phys. Chem. Lett. 2, 2241\u20132251 (2011).","journal-title":"J. Phys. Chem. Lett."},{"key":"2093_CR53","doi-asserted-by":"publisher","unstructured":"Honda, S., Shi, S. & Ueda, H. R. Smiles transformer: pre-trained molecular fingerprint for low data drug discovery. Preprint at https:\/\/doi.org\/10.48550\/arXiv.1911.04738 (2019).","DOI":"10.48550\/arXiv.1911.04738"},{"key":"2093_CR54","unstructured":"Hu, W. et al. Strategies for pre-training graph neural networks. In International Conference on Learning Representations. (OpenReview.net, 2020)."},{"key":"2093_CR55","doi-asserted-by":"publisher","first-page":"btaf275","DOI":"10.1093\/bioinformatics\/btaf275","volume":"41","author":"CE Lee","year":"2025","unstructured":"Lee, C. E., Kim, J. S., Min, J. H. & Han, S. W. SimSon: simple contrastive learning of SMILES for molecular property prediction. Bioinformatics. 41, btaf275 (2025).","journal-title":"Bioinformatics"},{"key":"2093_CR56","doi-asserted-by":"crossref","unstructured":"DeLong, E. R., DeLong, D. M. & Clarke-Pearson, D. L. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 44, 837\u2013845 (1988)","DOI":"10.2307\/2531595"},{"key":"2093_CR57","doi-asserted-by":"crossref","unstructured":"Zouhar, V. et al. A formal perspective on byte-pair encoding. In Findings of the Association for Computational Linguistics: ACL 2023. 598-614 (ACL, 2023).","DOI":"10.18653\/v1\/2023.findings-acl.38"},{"key":"2093_CR58","doi-asserted-by":"publisher","first-page":"432","DOI":"10.1038\/s42256-023-00639-z","volume":"5","author":"J Born","year":"2023","unstructured":"Born, J. & Manica, M. Regression transformer enables concurrent sequence regression and generation for molecular language modelling. Nat. Mach. Intell. 5, 432\u2013444 (2023).","journal-title":"Nat. Mach. Intell."},{"key":"2093_CR59","unstructured":"Ansari, A. F. et al. Chronos: learning the language of time series. Trans. Mach. Learn. Res. 2024(2024)"},{"key":"2093_CR60","doi-asserted-by":"publisher","first-page":"D1388","DOI":"10.1093\/nar\/gkaa971","volume":"49","author":"S Kim","year":"2021","unstructured":"Kim, S. et al. PubChem in 2021: new data content and improved web interfaces. Nucleic Acids Res. 49, D1388\u2013D1395 (2021).","journal-title":"Nucleic Acids Res."},{"key":"2093_CR61","unstructured":"Landrum, G. RDKit: Open-Source Cheminformatics Software, https:\/\/www.rdkit.org\/ (2016)."},{"key":"2093_CR62","doi-asserted-by":"publisher","first-page":"90","DOI":"10.1038\/nchem.1243","volume":"4","author":"GR Bickerton","year":"2012","unstructured":"Bickerton, G. R. et al. Quantifying the chemical beauty of drugs. Nat. Chem. 4, 90\u201398 (2012).","journal-title":"Nat. Chem."},{"key":"2093_CR63","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/1758-2946-1-8","volume":"1","author":"P Ertl","year":"2009","unstructured":"Ertl, P. & Schuffenhauer, A. Estimation of synthetic accessibility score of drug-like molecules based on molecular complexity and fragment contributions. J. Cheminform. 1, 1\u201311 (2009).","journal-title":"J. Cheminform."},{"key":"2093_CR64","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1023\/A:1011107422318","volume":"15","author":"R Mannhold","year":"2001","unstructured":"Mannhold, R. & Van de Waterbeemd, H. Substructure and whole molecule approaches for calculating log P. J. Comput. Aided Mol. Des. 15, 337\u2013354 (2001).","journal-title":"J. Comput. Aided Mol. Des."},{"key":"2093_CR65","unstructured":"Ramsundar, B., Eastman, P., Walters, P. & Pande, V. Deep learning for the life sciences: applying deep learning to genomics, microscopy, drug discovery, and more (O\u2019Reilly Media, 2019)."}],"container-title":["Communications Chemistry"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s42004-026-02093-5","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s42004-026-02093-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s42004-026-02093-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T22:48:50Z","timestamp":1789598930000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s42004-026-02093-5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,8]]},"references-count":65,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["2093"],"URL":"https:\/\/doi.org\/10.1038\/s42004-026-02093-5","relation":{},"ISSN":["2399-3669"],"issn-type":[{"value":"2399-3669","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,8]]},"assertion":[{"value":"23 November 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"298"}}