{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T08:21:24Z","timestamp":1777623684116,"version":"3.51.4"},"reference-count":66,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,4,17]],"date-time":"2023-04-17T00:00:00Z","timestamp":1681689600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,4,17]],"date-time":"2023-04-17T00:00:00Z","timestamp":1681689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"UK EPSRC Centre for Doctoral Training in Industrially Focused Mathematical Modelling","award":["EP\/L015803\/1"],"award-info":[{"award-number":["EP\/L015803\/1"]}]},{"name":"Lhasa Limited"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Cheminform"],"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Introduction and methodology<\/jats:title>\n                <jats:p>Pairs of similar compounds that only differ by a small structural modification but exhibit a large difference in their binding affinity for a given target are known as activity cliffs\u00a0(ACs). It has been hypothesised that QSAR models struggle to predict ACs and that ACs thus form a major source of prediction error. However, the AC-prediction power of modern QSAR methods and its quantitative relationship to general QSAR-prediction performance is still underexplored. We systematically construct nine distinct QSAR models by combining three molecular representation methods (extended-connectivity fingerprints, physicochemical-descriptor vectors and graph isomorphism networks) with three regression techniques (random forests, k-nearest neighbours and multilayer perceptrons); we then use each resulting model to classify pairs of similar compounds as ACs or non-ACs and to predict the activities of individual molecules in three case studies: dopamine receptor D2, factor Xa, and SARS-CoV-2 main protease.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results and conclusions<\/jats:title>\n                <jats:p>Our results provide strong support for the hypothesis that indeed QSAR models frequently fail to predict ACs. We observe low AC-sensitivity amongst the evaluated models when the activities of both compounds are unknown, but a substantial increase in AC-sensitivity when the actual activity of one of the compounds is given. Graph isomorphism features are found to be competitive with or superior to classical molecular representations for AC-classification and can thus be employed as baseline AC-prediction models or simple compound-optimisation tools. For general QSAR-prediction, however, extended-connectivity fingerprints still consistently deliver the best performance amongs the tested input representations. A potential future pathway to improve QSAR-modelling performance might be the development of techniques to increase AC-sensitivity.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Graphical Abstract<\/jats:title>\n                \n              <\/jats:sec>","DOI":"10.1186\/s13321-023-00708-w","type":"journal-article","created":{"date-parts":[[2023,4,17]],"date-time":"2023-04-17T10:08:55Z","timestamp":1681726135000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":51,"title":["Exploring QSAR models for activity-cliff prediction"],"prefix":"10.1186","volume":"15","author":[{"given":"Markus","family":"Dablander","sequence":"first","affiliation":[]},{"given":"Thierry","family":"Hanser","sequence":"additional","affiliation":[]},{"given":"Renaud","family":"Lambiotte","sequence":"additional","affiliation":[]},{"given":"Garrett M.","family":"Morris","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,4,17]]},"reference":[{"key":"708_CR1","unstructured":"Achdout H, Aimon A, Bar-David E, Barr H, Ben-Shmuel A, Bennett J, Bilenko VA, Bilenko VA, Boby ML, Borden B, Bowman GR, Brun J, et\u00a0al (2022) Open science discovery of oral non-covalent SARS-CoV-2 main protease inhibitor therapeutics. BioRxiv. https:\/\/www.biorxiv.org\/content\/early\/2022\/01\/30\/2020.10.29.339317. Accessed 19 Jan 2023"},{"key":"708_CR2","doi-asserted-by":"crossref","unstructured":"Akiba T, Sano S, Yanase T, Ohta T, Koyama M (2019) Optuna: a next-generation hyperparameter optimization framework. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 2623\u20132631","DOI":"10.1145\/3292500.3330701"},{"issue":"1","key":"708_CR3","doi-asserted-by":"publisher","first-page":"14710","DOI":"10.1038\/s41598-020-71696-2","volume":"10","author":"Y Asawa","year":"2020","unstructured":"Asawa Y, Yoshimori A, Bajorath J, Nakamura H (2020) Prediction of an MMP-1 inhibitor activity cliff using the SAR matrix approach and its experimental validation. Sci Rep 10(1):14710","journal-title":"Sci Rep"},{"issue":"6\u20137","key":"708_CR4","doi-asserted-by":"publisher","first-page":"438","DOI":"10.1002\/minf.201400026","volume":"33","author":"J Bajorath","year":"2014","unstructured":"Bajorath J (2014) Exploring activity cliffs from a chemoinformatics perspective. Mol Inf 33(6\u20137):438\u2013442","journal-title":"Mol Inf"},{"issue":"4","key":"708_CR5","doi-asserted-by":"publisher","first-page":"1226","DOI":"10.1021\/ci500012n","volume":"54","author":"JM Beck","year":"2014","unstructured":"Beck JM, Springer C (2014) Quantitative structure-activity relationship models of chemical transformations from matched pairs analyses. J Chem Inf Model 54(4):1226\u20131234","journal-title":"J Chem Inf Model"},{"issue":"1","key":"708_CR6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-020-00456-1","volume":"12","author":"AP Bento","year":"2020","unstructured":"Bento AP, Hersey A, F\u00e9lix E, Landrum G, Gaulton A, Atkinson F, Bellis LJ, de Veij M, Leach AR (2020) An open source chemical structure curation pipeline using RDKit. J Cheminformatics 12(1):1\u201316","journal-title":"J Cheminformatics"},{"key":"708_CR7","doi-asserted-by":"publisher","first-page":"898","DOI":"10.1039\/D2DD00077F","volume":"1","author":"H Chen","year":"2022","unstructured":"Chen H, Vogt M, Bajorath J (2022) DeepAC - conditional transformer-based chemical language model for the prediction of activity cliffs formed by bioactive compounds. Dig Discov 1:898\u2013909","journal-title":"Dig Discov"},{"key":"708_CR8","unstructured":"Chithrananda S, Grand G, Ramsundar B (2020) ChemBERTa: large-scale self-supervised pretraining for molecular property prediction. http:\/\/arxiv.org\/abs\/2010.09885"},{"issue":"8","key":"708_CR9","doi-asserted-by":"publisher","first-page":"1069","DOI":"10.1016\/j.drudis.2014.02.003","volume":"19","author":"M Cruz-Monteagudo","year":"2014","unstructured":"Cruz-Monteagudo M, Medina-Franco JL, P\u00e9rez-Castillo Y, Nicolotti O, Cordeiro MNDS, Borges F (2014) Activity cliffs in drug discovery: Dr Jekyll or Mr Hyde? Drug Discov Today 19(8):1069\u20131080","journal-title":"Drug Discov Today"},{"issue":"33","key":"708_CR10","doi-asserted-by":"publisher","first-page":"5043","DOI":"10.2174\/1381612822666160509124337","volume":"22","author":"M Cruz-Monteagudo","year":"2016","unstructured":"Cruz-Monteagudo M, Medina-Franco L, J, Perera-Sardi\u00f1a Y, Borges F, Tejera E, Paz-y Mino C, P\u00e9rez-Castillo Y, S\u00e1nchez-Rodr\u00edguez A, Contreras-Posada Z, Cordeiro ND, (2016) Probing the hypothesis of SAR continuity restoration by the removal of activity cliffs generators in QSAR. Curr Pharm Des 22(33):5043\u20135056","journal-title":"Curr Pharm Des"},{"key":"708_CR11","unstructured":"Dablander M, Lambiotte R, Morris GM, Hanser T (2021) Siamese neural networks work for activity cliff prediction. In: Poster presented at the 4th RSC-BMCS \/ RSC-CICAG Artificial Intelligence in Chemistry Symposium. https:\/\/www.researchgate.net\/publication\/362875964_Siamese_Neural_Networks_Work_for_Activity_Cliff_Prediction. Accessed 19 Jan 2023"},{"issue":"5","key":"708_CR12","doi-asserted-by":"publisher","first-page":"902","DOI":"10.1021\/acs.jcim.8b00173","volume":"58","author":"A Dalke","year":"2018","unstructured":"Dalke A, Hert J, Kramer C (2018) mmpdb: an open-source matched molecular pair platform for large multiproperty data sets. J Chem Inf Model 58(5):902\u2013910","journal-title":"J Chem Inf Model"},{"issue":"5","key":"708_CR13","doi-asserted-by":"publisher","first-page":"441","DOI":"10.1517\/17460441.2015.1019861","volume":"10","author":"D Dimova","year":"2015","unstructured":"Dimova D, Stumpfe D, Hu Y, Bajorath J (2015) Activity cliff clusters as a source of structure-activity relationship information. Expert Opin Drug Discov 10(5):441\u2013447","journal-title":"Expert Opin Drug Discov"},{"key":"708_CR14","unstructured":"Duvenaud DK, Maclaurin D, Iparraguirre J, Bombarell R, Hirzel T, Aspuru-Guzik A, Adams RP (2015) Convolutional networks on graphs for learning molecular fingerprints. In: Advances in Neural Information Processing Systems, pp 2224\u20132232"},{"key":"708_CR15","unstructured":"Fabian B, Edlich T, Gaspar H, Segler M, Meyers J, Fiscato M, Ahmed M (2020) Molecular representation learning with language models and domain-relevant auxiliary tasks. http:\/\/arxiv.org\/abs\/2011.13230"},{"key":"708_CR16","unstructured":"Fey M, Lenssen JE (2019) Fast graph representation learning with PyTorch Geometric. http:\/\/arxiv.org\/abs\/1903.02428"},{"key":"708_CR17","unstructured":"Gilmer J, Schoenholz SS, Riley PF, Vinyals O, Dahl GE (2017) Neural message passing for quantum chemistry. In: International Conference on Machine Learning, PMLR, pp 1263\u20131272"},{"issue":"1","key":"708_CR18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1021\/ci400572x","volume":"54","author":"A Golbraikh","year":"2014","unstructured":"Golbraikh A, Muratov E, Fourches D, Tropsha A (2014) Data set modelability by QSAR. J Chem Inf Model 54(1):1\u20134","journal-title":"J Chem Inf Model"},{"issue":"9","key":"708_CR19","doi-asserted-by":"publisher","first-page":"2354","DOI":"10.1021\/ci300306a","volume":"52","author":"K Heikamp","year":"2012","unstructured":"Heikamp K, Hu X, Yan A, Bajorath J (2012) Prediction of activity cliffs using support vector machines. J Chem Inf Model 52(9):2354\u20132365","journal-title":"J Chem Inf Model"},{"issue":"2","key":"708_CR20","doi-asserted-by":"publisher","first-page":"253","DOI":"10.1142\/S0218213011000140","volume":"20","author":"F Hoonakker","year":"2011","unstructured":"Hoonakker F, Lachiche N, Varnek A, Wagner A (2011) Condensed graph of reaction: considering a chemical reaction as one single pseudo molecule. Int J Artif Intell Tools 20(2):253\u2013270","journal-title":"Int J Artif Intell Tools"},{"issue":"9","key":"708_CR21","doi-asserted-by":"publisher","first-page":"1631","DOI":"10.1021\/acs.jcim.6b00359","volume":"56","author":"D Horvath","year":"2016","unstructured":"Horvath D, Marcou G, Varnek A, Kayastha S, de la Vega de Le\u00f3n A, Bajorath J, (2016) Prediction of activity cliffs using condensed graphs of reaction representations. J Chem Inf Model 56(9):1631\u20131640","journal-title":"J Chem Inf Model"},{"key":"708_CR22","unstructured":"Hu W, Liu B, Gomes J, Zitnik M, Liang P, Pande V, Leskovec J (2019) Strategies for pre-training graph neural networks. http:\/\/arxiv.org\/abs\/1905.12265"},{"issue":"7","key":"708_CR23","doi-asserted-by":"publisher","first-page":"1806","DOI":"10.1021\/ci300274c","volume":"52","author":"Y Hu","year":"2012","unstructured":"Hu Y, Bajorath J (2012) Extending the activity cliff concept: structural categorization of activity cliffs and systematic identification of different types of cliffs in the ChEMBL database. J Chem Inf Model 52(7):1806\u20131811","journal-title":"J Chem Inf Model"},{"issue":"5","key":"708_CR24","doi-asserted-by":"publisher","first-page":"1062","DOI":"10.1021\/ci500742b","volume":"55","author":"J Husby","year":"2015","unstructured":"Husby J, Bottegoni G, Kufareva I, Abagyan R, Cavalli A (2015) Structure-based predictions of activity cliffs. J Chem Inf Model 55(5):1062\u20131076","journal-title":"J Chem Inf Model"},{"key":"708_CR25","unstructured":"Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. In: Proceedings of Machine Learning Research, pp 448\u2013456"},{"key":"708_CR26","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.1007\/s10822-021-00380-y","volume":"35","author":"J Iqbal","year":"2021","unstructured":"Iqbal J, Vogt M, Bajorath J (2021) Prediction of activity cliffs on the basis of images using convolutional neural networks. J Comput Aided Mol Des 35:1157\u20131164","journal-title":"J Comput Aided Mol Des"},{"issue":"6","key":"708_CR27","doi-asserted-by":"publisher","first-page":"335","DOI":"10.1016\/0898-5529(90)90060-L","volume":"3","author":"P Jauffret","year":"1990","unstructured":"Jauffret P, Tonnelier C, Hanser T, Kaufmann G, Wolff R (1990) Machine learning of generic reactions: 2. Toward an advanced computer representation of chemical reactions. Tetrahedron Comput Methodol 3(6):335\u2013349","journal-title":"Tetrahedron Comput Methodol"},{"issue":"1","key":"708_CR28","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-020-00479-8","volume":"13","author":"D Jiang","year":"2021","unstructured":"Jiang D, Wu Z, Hsieh CY, Chen G, Liao B, Wang Z, Shen C, Cao D, Wu J, Hou T (2021) Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models. J Cheminformatics 13(1):1\u201323","journal-title":"J Cheminformatics"},{"key":"708_CR29","doi-asserted-by":"publisher","first-page":"271","DOI":"10.1002\/3527603743.ch11","volume":"23","author":"PW Kenny","year":"2005","unstructured":"Kenny PW, Sadowski J (2005) Structure modification in chemical databases. Chemoinformatics Drug Discov 23:271\u2013285","journal-title":"Chemoinformatics Drug Discov"},{"issue":"6","key":"708_CR30","first-page":"2431","volume":"13","author":"MR Keyvanpour","year":"2021","unstructured":"Keyvanpour MR, Barani Shirzad M, Moradi F (2021) PCAC: a new method for predicting compounds with activity cliff property in QSAR approach. Int J Inf Technol 13(6):2431\u20132437","journal-title":"Int J Inf Technol"},{"key":"708_CR31","unstructured":"Kipf TN, Welling M (2016) Semi-supervised classification with graph convolutional networks. arXiv preprint. https:\/\/arxiv.org\/abs\/1609.02907 [cs.LG]"},{"key":"708_CR32","unstructured":"Landrum G (2006) RDKit: open-source cheminformatics"},{"issue":"2","key":"708_CR33","doi-asserted-by":"publisher","first-page":"151","DOI":"10.2174\/1568026013395380","volume":"1","author":"J Leadley","year":"2001","unstructured":"Leadley J (2001) Coagulation factor Xa inhibition: biological background and rationale. Curr Top Med Chem 1(2):151\u2013159","journal-title":"Curr Top Med Chem"},{"issue":"10","key":"708_CR34","doi-asserted-by":"publisher","first-page":"2654","DOI":"10.1021\/ci5003944","volume":"54","author":"De la Vega","year":"2014","unstructured":"la Vega De, de Le\u00f3n A, Bajorath J (2014) Prediction of compound potency changes in matched molecular pairs using support vector regression. J Chem Inf Model 54(10):2654\u20132663","journal-title":"J Chem Inf Model"},{"key":"708_CR35","doi-asserted-by":"publisher","first-page":"D198","DOI":"10.1093\/nar\/gkl999","volume":"35","author":"T Liu","year":"2007","unstructured":"Liu T, Lin Y, Wen X, Jorissen RN, Gilson MK (2007) BindingDB: a web-accessible database of experimentally determined protein-ligand binding affinities. Nucleic Acids Res 35:D198\u2013D201","journal-title":"Nucleic Acids Res"},{"key":"708_CR36","unstructured":"Loshchilov I, Hutter F (2017) Decoupled weight decay regularization. http:\/\/arxiv.org\/abs\/1711.05101"},{"issue":"4","key":"708_CR37","doi-asserted-by":"publisher","first-page":"1535","DOI":"10.1021\/ci060117s","volume":"46","author":"GM Maggiora","year":"2006","unstructured":"Maggiora GM (2006) On outliers and activity cliffs: why QSAR often disappoints. J Chem Inf Model 46(4):1535\u20131535","journal-title":"J Chem Inf Model"},{"issue":"24","key":"708_CR38","doi-asserted-by":"publisher","first-page":"5441","DOI":"10.1039\/C8SC00148K","volume":"9","author":"A Mayr","year":"2018","unstructured":"Mayr A, Klambauer G, Unterthiner T, Steijaert M, Wegner JK, Ceulemans H, Clevert DA, Hochreiter S (2018) Large-scale comparison of machine learning methods for drug target prediction on ChEMBL. Chem Sci 9(24):5441\u20135451","journal-title":"Chem Sci"},{"issue":"5","key":"708_CR39","doi-asserted-by":"publisher","first-page":"553","DOI":"10.1111\/cbdd.12115","volume":"81","author":"JL Medina-Franco","year":"2013","unstructured":"Medina-Franco JL (2013) Activity cliffs: facts or artifacts? Chem Biol Drug Design 81(5):553\u2013556","journal-title":"Chem Biol Drug Design"},{"issue":"2","key":"708_CR40","doi-asserted-by":"publisher","first-page":"664","DOI":"10.1021\/acs.jcim.0c01208","volume":"61","author":"J Menke","year":"2021","unstructured":"Menke J, Koch O (2021) Using domain-specific fingerprints generated through neural networks to enhance ligand-based virtual screening. J Chem Inf Model 61(2):664\u2013675","journal-title":"J Chem Inf Model"},{"issue":"4","key":"708_CR41","doi-asserted-by":"publisher","first-page":"927","DOI":"10.1021\/ci3000503","volume":"52","author":"V Namasivayam","year":"2012","unstructured":"Namasivayam V, Bajorath J (2012) Searching for coordinated activity cliffs using particle swarm optimization. J Chem Inf Model 52(4):927\u2013934","journal-title":"J Chem Inf Model"},{"issue":"12","key":"708_CR42","doi-asserted-by":"publisher","first-page":"3131","DOI":"10.1021\/ci400597d","volume":"53","author":"V Namasivayam","year":"2013","unstructured":"Namasivayam V, Iyer P, Bajorath J (2013) Prediction of individual compounds forming activity cliffs using emerging chemical patterns. J Chem Inf Model 53(12):3131\u20133139","journal-title":"J Chem Inf Model"},{"issue":"10","key":"708_CR43","doi-asserted-by":"publisher","first-page":"2341","DOI":"10.1021\/acs.jcim.2c00327","volume":"62","author":"J Park","year":"2022","unstructured":"Park J, Sung G, Lee S, Kang S, Park C (2022) ACGCN: graph convolutional networks for activity cliff prediction between matched molecular pairs. J Chem Inf Model 62(10):2341\u20132351. https:\/\/doi.org\/10.1021\/acs.jcim.2c00327","journal-title":"J Chem Inf Model"},{"key":"708_CR44","unstructured":"Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, Desmaison A, Kopf A, Yang E, DeVito Z, Raison M, Tejani A, Chilamkurthy S, Steiner B, Fang L, Bai J, Chintala S (2019) PyTorch: an imperative style, high-performance deep learning library. In: Wallach H, Larochelle H, Beygelzimer A, d\u2019 Alch\u00e9-Buc F, Fox E, Garnett R (eds) Advances in Neural Information Processing Systems, Curran Associates, Inc., vol\u00a032. https:\/\/proceedings.neurips.cc\/paper\/2019\/file\/bdbca288fee7f92f2bfa9f7012727740-Paper.pdf. Accessed 19 Jan 2023"},{"key":"708_CR45","first-page":"2825","volume":"12","author":"F Pedregosa","year":"2011","unstructured":"Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V et al (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825\u20132830","journal-title":"J Mach Learn Res"},{"issue":"3","key":"708_CR46","doi-asserted-by":"publisher","first-page":"1884","DOI":"10.1021\/acs.jctc.8b01290","volume":"15","author":"L P\u00e9rez-Benito","year":"2019","unstructured":"P\u00e9rez-Benito L, Casajuana-Martin N, Jim\u00e9nez-Ros\u00e9s M, van Vlijmen H, Tresadern G (2019) Predicting activity cliffs with free-energy perturbation. J Chem Theory Comput 15(3):1884\u20131895","journal-title":"J Chem Theory Comput"},{"issue":"5","key":"708_CR47","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1021\/ci100050t","volume":"50","author":"D Rogers","year":"2010","unstructured":"Rogers D, Hahn M (2010) Extended-connectivity fingerprints. J Chem Inf Model 50(5):742\u2013754","journal-title":"J Chem Inf Model"},{"key":"708_CR48","doi-asserted-by":"crossref","unstructured":"Sabando MV, Ponzoni I, Milios EE, Soto AJ (2021) Using molecular embeddings in QSAR modeling: does it make a difference? http:\/\/arxiv.org\/abs\/2104.02604","DOI":"10.1093\/bib\/bbab365"},{"issue":"2","key":"708_CR49","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1002\/syn.890010203","volume":"1","author":"P Seeman","year":"1987","unstructured":"Seeman P (1987) Dopamine receptors and the dopamine hypothesis of schizophrenia. Synapse 1(2):133\u2013152","journal-title":"Synapse"},{"issue":"4","key":"708_CR50","doi-asserted-by":"publisher","first-page":"1969","DOI":"10.1021\/acs.jcim.9b01067","volume":"60","author":"RP Sheridan","year":"2020","unstructured":"Sheridan RP, Karnachi P, Tudor M, Xu Y, Liaw A, Shah F, Cheng AC, Joshi E, Glick M, Alvarez J (2020) Experimental error, kurtosis, activity cliffs, and methodology: what limits the predictivity of quantitative structure-activity relationship models. J Chem Inf Model 60(4):1969\u20131982","journal-title":"J Chem Inf Model"},{"key":"708_CR51","unstructured":"Silipo C, Vittoria A (1991) QSAR, rational approaches to the design of bioactive compounds. In: Proceedings of European Symposium on Quantitative Structure-Activity Relationships, Distributors for the US and Canada, Elsevier Science"},{"issue":"1","key":"708_CR52","first-page":"1929","volume":"15","author":"N Srivastava","year":"2014","unstructured":"Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929\u20131958","journal-title":"J Mach Learn Res"},{"issue":"104","key":"708_CR53","first-page":"197","volume":"130","author":"T Stepi\u0161nik","year":"2021","unstructured":"Stepi\u0161nik T, \u0160krlj B, Wicker J, Kocev D (2021) A comprehensive comparison of molecular feature representations for use in predictive modeling. Comput Biol Med 130(104):197","journal-title":"Comput Biol Med"},{"issue":"1","key":"708_CR54","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1021\/jm401120g","volume":"57","author":"D Stumpfe","year":"2014","unstructured":"Stumpfe D, Hu Y, Dimova D, Bajorath J (2014) Recent progress in understanding activity cliffs and their utility in medicinal chemistry: miniperspective. J Med Chem 57(1):18\u201328","journal-title":"J Med Chem"},{"issue":"11","key":"708_CR55","doi-asserted-by":"publisher","first-page":"14360","DOI":"10.1021\/acsomega.9b02221","volume":"4","author":"D Stumpfe","year":"2019","unstructured":"Stumpfe D, Hu H, Bajorath J (2019) Evolving concept of activity cliffs. ACS Omega 4(11):14360\u201314368","journal-title":"ACS Omega"},{"issue":"9","key":"708_CR56","doi-asserted-by":"publisher","first-page":"929","DOI":"10.1007\/s10822-020-00315-z","volume":"34","author":"D Stumpfe","year":"2020","unstructured":"Stumpfe D, Hu H, Bajorath J (2020) Advances in exploring activity cliffs. J Comput Aided Mol Des 34(9):929\u2013942","journal-title":"J Comput Aided Mol Des"},{"key":"708_CR57","doi-asserted-by":"publisher","DOI":"10.1002\/minf.202000103","author":"S Tamura","year":"2020","unstructured":"Tamura S, Miyao T, Funatsu K (2020) Ligand-based activity cliff prediction models with applicability domain. Mol Inform. https:\/\/doi.org\/10.1002\/minf.202000103","journal-title":"Mol Inform"},{"key":"708_CR58","volume-title":"Handbook of molecular descriptors","author":"R Todeschini","year":"2008","unstructured":"Todeschini R, Consonni V (2008) Handbook of molecular descriptors. John Wiley & Sons, New York"},{"issue":"17","key":"708_CR59","doi-asserted-by":"publisher","DOI":"10.1016\/j.bmcl.2020.127377","volume":"30","author":"S Ullrich","year":"2020","unstructured":"Ullrich S, Nitsche C (2020) The SARS-CoV-2 main protease as drug target. Bioorg Med Chem Lett 30(17):127377","journal-title":"Bioorg Med Chem Lett"},{"key":"708_CR60","doi-asserted-by":"crossref","unstructured":"Van\u00a0Tilborg D, Alenicheva A, Grisoni F (2022) Exposing the limitations of molecular machine learning with activity cliffs. ChemRxiv. https:\/\/chemrxiv.org\/engage\/chemrxiv\/article-details\/623de3fbab0051148698fbcf. Accessed 19 Jan 2023","DOI":"10.26434\/chemrxiv-2022-mfq52-v2"},{"key":"708_CR61","unstructured":"Veli\u010dkovi\u0107 P, Cucurull G, Casanova A, Romero A, Lio P, Bengio Y (2017) Graph attention networks. http:\/\/arxiv.org\/abs\/1710.10903"},{"issue":"8","key":"708_CR62","doi-asserted-by":"publisher","first-page":"1848","DOI":"10.1021\/ci2002473","volume":"51","author":"M Vogt","year":"2011","unstructured":"Vogt M, Huang Y, Bajorath J (2011) From activity cliffs to activity ridges: informative data structures for SAR analysis. J Chem Inf Model 51(8):1848\u20131856","journal-title":"J Chem Inf Model"},{"key":"708_CR63","unstructured":"Wang Y, Wang J, Cao Z, Farimani AB (2021) MolCLR: molecular contrastive learning of representations via graph neural networks. http:\/\/arxiv.org\/abs\/2102.10056"},{"issue":"1\u20132","key":"708_CR64","doi-asserted-by":"publisher","first-page":"1600118","DOI":"10.1002\/minf.201600118","volume":"36","author":"DA Winkler","year":"2017","unstructured":"Winkler DA, Le TC (2017) Performance of deep and shallow neural networks, the universal approximation theorem, activity cliffs, and QSAR. Mol Inform 36(1\u20132):1600118","journal-title":"Mol Inform"},{"issue":"6","key":"708_CR65","doi-asserted-by":"publisher","first-page":"1692","DOI":"10.1039\/C8SC04175J","volume":"10","author":"R Winter","year":"2019","unstructured":"Winter R, Montanari F, No\u00e9 F, Clevert DA (2019) Learning continuous and data-driven molecular descriptors by translating equivalent chemical representations. Chem Sci 10(6):1692\u20131701","journal-title":"Chem Sci"},{"key":"708_CR66","unstructured":"Xu K, Hu W, Leskovec J, Jegelka S (2018) How powerful are graph neural networks? http:\/\/arxiv.org\/abs\/1810.00826"}],"container-title":["Journal of Cheminformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13321-023-00708-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13321-023-00708-w\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13321-023-00708-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,4,17]],"date-time":"2023-04-17T10:16:27Z","timestamp":1681726587000},"score":1,"resource":{"primary":{"URL":"https:\/\/jcheminf.biomedcentral.com\/articles\/10.1186\/s13321-023-00708-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,4,17]]},"references-count":66,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["708"],"URL":"https:\/\/doi.org\/10.1186\/s13321-023-00708-w","relation":{},"ISSN":["1758-2946"],"issn-type":[{"value":"1758-2946","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,4,17]]},"assertion":[{"value":"26 January 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"10 March 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 April 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"47"}}