{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"Research Square"}],"indexed":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T07:02:36Z","timestamp":1784098956232,"version":"3.55.0"},"posted":{"date-parts":[[2024,12,4]]},"group-title":"In Review","reference-count":21,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2024,12,4]],"date-time":"2024-12-04T00:00:00Z","timestamp":1733270400000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2023,5,27]]},"abstract":"<title>Abstract<\/title>\n                <p>Modeling molecular activity and quantitative structure-activity relationships of chemical compounds is critical in drug design. Graph neural networks, which utilize molecular structures as frames, have shown success in assessing the biological activity of chemical compounds, guiding the selection and optimization of candidates for further development. However, current models often overlook activity cliffs (ACs)\u2014cases where structurally similar molecules exhibit different bioactivities\u2014due to latent spaces primarily optimized for structural features. Here, we introduce AC-awareness (ACA), an inductive bias designed to enhance molecular representation learning for activity modeling. The ACA jointly optimizes metric learning in the latent space and task performance in the target space, making models more sensitive to ACs. We develop ACANET, an AC-informed contrastive learning approach that can be integrated with any graph neural network. Experiments on 39 benchmark datasets demonstrate that AC-informed representations of chemical compounds consistently outperform standard models in bioactivity prediction across both regression and classification tasks. AC-informedmodels show strong performance in predicting pharmacokinetic and safety-relevant molecular properties. ACA paves the way toward activity-informed molecular representations, providing a valuable tool for the early stages of lead compound identification, refinement, and virtual screening.<\/p>","DOI":"10.21203\/rs.3.rs-2988283\/v2","type":"posted-content","created":{"date-parts":[[2024,12,4]],"date-time":"2024-12-04T08:28:16Z","timestamp":1733300896000},"source":"Crossref","is-referenced-by-count":2,"title":["Activity Cliff-Informed Contrastive Learning for Molecular Property Prediction"],"prefix":"10.21203","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7114-3664","authenticated-orcid":false,"given":"Wanxiang","family":"Shen","sequence":"first","affiliation":[{"name":"Department of Biomedical Informatics, Harvard Medical School"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chao","family":"Cui","sequence":"additional","affiliation":[{"name":"Tsinghua Shenzhen International Graduate School, Tsinghua University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaorui","family":"Su","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, Harvard Medical School"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zaixi","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, Harvard Medical School"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alejandro Velez","family":"Arce","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, Harvard Medical School"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianming","family":"Wang","sequence":"additional","affiliation":[{"name":"Interdisciplinary Graduate Program in Integrative Biotechnology, Yonsei University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiangcheng","family":"Shi","sequence":"additional","affiliation":[{"name":"Department of Chemistry, National University of Singapore"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanbing","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Chemistry, National University of Singapore"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jie","family":"Wu","sequence":"additional","affiliation":[{"name":"Department of Chemistry, National University of Singapore"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5473-8022","authenticated-orcid":false,"given":"Yu Zong","family":"Chen","sequence":"additional","affiliation":[{"name":"Tsinghua University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8530-7228","authenticated-orcid":false,"given":"Marinka","family":"Zitnik","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, Harvard Medical School"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","reference":[{"key":"ref1","doi-asserted-by":"crossref","first-page":"673","DOI":"10.1038\/s41586-023-05905-z","article-title":"Computational approaches streamlining drug discovery","volume":"616","author":"Sadybekov AV","year":"2023","unstructured":"Sadybekov, A. V. & Katritch, V. Computational approaches streamlining drug discovery. Nature 616, 673\u2013685 (2023).","journal-title":"Nature"},{"key":"ref2","doi-asserted-by":"crossref","first-page":"948","DOI":"10.1038\/nrd4128","article-title":"Chemical predictive modelling to improve compound quality","volume":"12","author":"Cumming JG","year":"2013","unstructured":"Cumming, J. G., Davis, A. M., Muresan, S., Haeberlein, M. & Chen, H. Chemical predictive modelling to improve compound quality. Nature reviews Drug discovery 12, 948\u2013962 (2013).","journal-title":"Nature reviews Drug discovery"},{"key":"ref3","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1038\/s42256-022-00447-x","article-title":"Barati Farimani, A. Molecular contrastive learning of representations via graph neural networks","volume":"4","author":"Wang Y","year":"2022","unstructured":"Wang, Y., Wang, J., Cao, Z. & Barati Farimani, A. Molecular contrastive learning of representations via graph neural networks. Nature Machine Intelligence 4, 279\u2013287 (2022).","journal-title":"Nature Machine Intelligence"},{"key":"ref4","doi-asserted-by":"crossref","first-page":"334","DOI":"10.1038\/s42256-021-00301-6","article-title":"Out-of-the-box deep learning prediction of pharmaceutical properties by broadly learned knowledge-based molecular representations","volume":"3","author":"Shen WX","year":"2021","unstructured":"Shen, W. X. et al. Out-of-the-box deep learning prediction of pharmaceutical properties by broadly learned knowledge-based molecular representations. Nature Machine Intelligence 3, 334\u2013343 (2021).","journal-title":"Nature Machine Intelligence"},{"key":"ref5","doi-asserted-by":"crossref","first-page":"3521","DOI":"10.1038\/s41467-021-23720-w","article-title":"Algebraic graph-assisted bidirectional transformers for molecular property prediction","volume":"12","author":"Chen D","year":"2021","unstructured":"Chen, D. et al. Algebraic graph-assisted bidirectional transformers for molecular property prediction. Nature communications 12, 3521 (2021).","journal-title":"Nature communications"},{"key":"ref6","doi-asserted-by":"crossref","first-page":"645","DOI":"10.1038\/s42256-022-00501-8","article-title":"An adaptive graph learning method for automated molecular interactions and properties predictions","volume":"4","author":"Li Y","year":"2022","unstructured":"Li, Y. et al. An adaptive graph learning method for automated molecular interactions and properties predictions. Nature Machine Intelligence 4, 645\u2013651 (2022).","journal-title":"Nature Machine Intelligence"},{"key":"ref7","doi-asserted-by":"crossref","first-page":"114","DOI":"10.1038\/s41467-022-35692-6","article-title":"Leveraging molecular structure and bioactivity with chemical language models for de novo drug design","volume":"14","author":"Moret M","year":"2023","unstructured":"Moret, M. et al. Leveraging molecular structure and bioactivity with chemical language models for de novo drug design. Nature Communications 14, 114 (2023).","journal-title":"Nature Communications"},{"key":"ref8","first-page":"1","article-title":"Accurate prediction of molecular properties and drug targets using a self-supervised image representation learning framework","author":"Zeng X","year":"2022","unstructured":"Zeng, X. et al. Accurate prediction of molecular properties and drug targets using a self-supervised image representation learning framework. Nature Machine Intelligence, 1\u201313 (2022).","journal-title":"Nature Machine Intelligence"},{"key":"ref9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13321-020-00479-8","article-title":"Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models","volume":"13","author":"Jiang D","year":"2021","unstructured":"Jiang, D. et al. Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models. Journal of cheminformatics 13, 1\u201323 (2021).","journal-title":"Journal of cheminformatics"},{"key":"ref10","doi-asserted-by":"crossref","first-page":"5938","DOI":"10.1021\/acs.jcim.2c01073","article-title":"Exposing the limitations of molecular machine learning with activity cliffs","volume":"62","author":"Tilborg D","year":"2022","unstructured":"van Tilborg, D., Alenicheva, A. & Grisoni, F. Exposing the limitations of molecular machine learning with activity cliffs. Journal of Chemical Information and Modeling 62, 5938\u20135951 (2022).","journal-title":"Journal of Chemical Information and Modeling"},{"key":"ref11","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1186\/s13321-023-00708-w","article-title":"Exploring QSAR models for activity-cliff prediction","volume":"15","author":"Dablander M","year":"2023","unstructured":"Dablander, M., Hanser, T., Lambiotte, R. & Morris, G. M. Exploring QSAR models for activity-cliff prediction. Journal of Cheminformatics 15, 47 (2023).","journal-title":"Journal of Cheminformatics"},{"key":"ref12","author":"Zha K","year":"2022","unstructured":"Zha, K., Cao, P., Yang, Y. & Katabi, D. Supervised Contrastive Regression. arXiv preprint arXiv:2210.01189 (2022)."},{"key":"ref13","doi-asserted-by":"crossref","first-page":"15494","DOI":"10.1021\/acs.jmedchem.0c01608","article-title":"Fragment-to-lead medicinal chemistry publications in 2019","volume":"63","author":"Jahnke W","year":"2020","unstructured":"Jahnke, W. et al. Fragment-to-lead medicinal chemistry publications in 2019. Journal of medicinal chemistry 63, 15494\u201315507 (2020).","journal-title":"Journal of medicinal chemistry"},{"key":"ref14","doi-asserted-by":"crossref","first-page":"D1100-D1107","DOI":"10.1093\/nar\/gkr777","article-title":"ChEMBL: a large-scale bioactivity database for drug discovery","volume":"40","author":"Gaulton A","year":"2012","unstructured":"Gaulton, A. et al. ChEMBL: a large-scale bioactivity database for drug discovery. Nucleic acids research 40, D1100-D1107 (2012).","journal-title":"Nucleic acids research"},{"key":"ref15","doi-asserted-by":"crossref","first-page":"D1083-D1090","DOI":"10.1093\/nar\/gkt1031","article-title":"The ChEMBL bioactivity database: an update","volume":"42","author":"Bento AP","year":"2014","unstructured":"Bento, A. P. et al. The ChEMBL bioactivity database: an update. Nucleic acids research 42, D1083-D1090 (2014).","journal-title":"Nucleic acids research"},{"key":"ref16","doi-asserted-by":"crossref","first-page":"2932","DOI":"10.1021\/jm201706b","article-title":"r. Exploring activity cliffs in medicinal chemistry: miniperspective","volume":"55","author":"Stumpfe D","year":"2012","unstructured":"Stumpfe, D. & Bajorath, J. r. Exploring activity cliffs in medicinal chemistry: miniperspective. Journal of medicinal chemistry 55, 2932\u20132942 (2012).","journal-title":"Journal of medicinal chemistry"},{"key":"ref17","author":"Schroff F","year":"2015","unstructured":"Schroff, F., Kalenichenko, D. & Philbin, J. Facenet: A unified embedding for face recognition and clustering. in Proceedings of the IEEE conference on computer vision and pattern recognition. 815\u2013823 (2015)."},{"key":"ref18","first-page":"13260","article-title":"Principal neighbourhood aggregation for graph nets","volume":"33","author":"Corso G","year":"2020","unstructured":"Corso, G., Cavalleri, L., Beaini, D., Li\u00f2, P. & Veli\u010dkovi\u0107, P. Principal neighbourhood aggregation for graph nets. Advances in Neural Information Processing Systems 33, 13260\u201313271 (2020).","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref19","doi-asserted-by":"crossref","first-page":"8749","DOI":"10.1021\/acs.jmedchem.9b00959","article-title":"Pushing the boundaries of molecular representation for drug discovery with the graph attention mechanism","volume":"63","author":"Xiong Z","year":"2019","unstructured":"Xiong, Z. et al. Pushing the boundaries of molecular representation for drug discovery with the graph attention mechanism. Journal of medicinal chemistry 63, 8749\u20138760 (2019).","journal-title":"Journal of medicinal chemistry"},{"key":"ref20","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13321-020-0416-x","article-title":"Visualization of very large high-dimensional data sets as minimum spanning trees","volume":"12","author":"Probst D","year":"2020","unstructured":"Probst, D. & Reymond, J.-L. Visualization of very large high-dimensional data sets as minimum spanning trees. Journal of Cheminformatics 12, 1\u201313 (2020).","journal-title":"Journal of Cheminformatics"},{"key":"ref21","author":"McInnes L","year":"2018","unstructured":"McInnes, L., Healy, J. & Melville, J. Umap: Uniform manifold approximation and projection for dimension reduction. arXiv preprint arXiv:1802.03426 (2018)."}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/www.researchsquare.com\/article\/rs-2988283\/v2","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.researchsquare.com\/article\/rs-2988283\/v2.html","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T18:49:09Z","timestamp":1774378149000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.researchsquare.com\/article\/rs-2988283\/v2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12,4]]},"references-count":21,"URL":"https:\/\/doi.org\/10.21203\/rs.3.rs-2988283\/v2","relation":{"is-version-of":[{"id-type":"doi","id":"10.21203\/rs.3.rs-2988283\/v1","asserted-by":"subject"},{"id-type":"doi","id":"10.21203\/rs.3.rs-2988283\/v3","asserted-by":"subject"}],"has-version":[{"id-type":"doi","id":"10.21203\/rs.3.rs-2988283\/v3","asserted-by":"object"},{"id-type":"doi","id":"10.21203\/rs.3.rs-2988283\/v1","asserted-by":"object"}]},"subject":[],"published":{"date-parts":[[2024,12,4]]},"subtype":"preprint"}}