{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T19:42:39Z","timestamp":1775245359487,"version":"3.50.1"},"reference-count":72,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2021,2,17]],"date-time":"2021-02-17T00:00:00Z","timestamp":1613520000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,2,17]],"date-time":"2021-02-17T00:00:00Z","timestamp":1613520000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Key R&D Program of Zhejiang Province","award":["2020C03010"],"award-info":[{"award-number":["2020C03010"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["21575128"],"award-info":[{"award-number":["21575128"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["81773632"],"award-info":[{"award-number":["81773632"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100004731","name":"Natural Science Foundation of Zhejiang Province","doi-asserted-by":"publisher","award":["LZ19H300001"],"award-info":[{"award-number":["LZ19H300001"]}],"id":[{"id":"10.13039\/501100004731","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Cheminform"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Graph neural networks (GNN) has been considered as an attractive modelling method for molecular property prediction, and numerous studies have shown that GNN could yield more promising results than traditional descriptor-based methods. In this study, based on 11 public datasets covering various property endpoints, the predictive capacity and computational efficiency of the prediction models developed by eight machine learning (ML) algorithms, including four descriptor-based models (SVM, XGBoost, RF and DNN) and four graph-based models (GCN, GAT, MPNN and Attentive FP), were extensively tested and compared. The results demonstrate that on average the descriptor-based models outperform the graph-based models in terms of prediction accuracy and computational efficiency. SVM generally achieves the best predictions for the regression tasks. Both RF and XGBoost can achieve reliable predictions for the classification tasks, and some of the graph-based models, such as Attentive FP and GCN, can yield outstanding performance for a fraction of larger or multi-task datasets. In terms of computational cost, XGBoost and RF are the two most efficient algorithms and only need a few seconds to train a model even for a large dataset. The model interpretations by the SHAP method can effectively explore the established domain knowledge for the descriptor-based models. Finally, we explored use of these models for virtual screening (VS) towards HIV and demonstrated that different ML algorithms offer diverse VS profiles. All in all, we believe that the off-the-shelf descriptor-based models still can be directly employed to accurately predict various chemical endpoints with excellent computability and interpretability.<\/jats:p>","DOI":"10.1186\/s13321-020-00479-8","type":"journal-article","created":{"date-parts":[[2021,2,18]],"date-time":"2021-02-18T03:28:52Z","timestamp":1613618932000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":493,"title":["Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models"],"prefix":"10.1186","volume":"13","author":[{"given":"Dejun","family":"Jiang","sequence":"first","affiliation":[]},{"given":"Zhenxing","family":"Wu","sequence":"additional","affiliation":[]},{"given":"Chang-Yu","family":"Hsieh","sequence":"additional","affiliation":[]},{"given":"Guangyong","family":"Chen","sequence":"additional","affiliation":[]},{"given":"Ben","family":"Liao","sequence":"additional","affiliation":[]},{"given":"Zhe","family":"Wang","sequence":"additional","affiliation":[]},{"given":"Chao","family":"Shen","sequence":"additional","affiliation":[]},{"given":"Dongsheng","family":"Cao","sequence":"additional","affiliation":[]},{"given":"Jian","family":"Wu","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7227-2580","authenticated-orcid":false,"given":"Tingjun","family":"Hou","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2021,2,17]]},"reference":[{"key":"479_CR1","doi-asserted-by":"publisher","first-page":"497","DOI":"10.2174\/138620709788489082","volume":"12","author":"T Hou","year":"2009","unstructured":"Hou T, Li Y, Zhang W et al (2009) Recent developments of in silico predictions of intestinal absorption and oral bioavailability. Comb Chem High Throughput Screening 12:497\u2013506","journal-title":"Comb Chem High Throughput Screening"},{"key":"479_CR2","doi-asserted-by":"publisher","first-page":"624","DOI":"10.1016\/j.tips.2019.07.005","volume":"40","author":"AO Basile","year":"2019","unstructured":"Basile AO, Yahi A, Tatonetti NP (2019) Artificial intelligence for drug toxicity and safety. Trends Pharmacol Sci 40:624\u2013635","journal-title":"Trends Pharmacol Sci"},{"key":"479_CR3","doi-asserted-by":"publisher","first-page":"4463","DOI":"10.1021\/jm0303195","volume":"47","author":"XY Xia","year":"2004","unstructured":"Xia XY, Maliski EG, Gallant P et al (2004) Classification of kinase inhibitors using a Bayesian model. J Med Chem 47:4463\u20134470","journal-title":"J Med Chem"},{"key":"479_CR4","doi-asserted-by":"publisher","first-page":"2875","DOI":"10.1021\/mp300198d","volume":"9","author":"S Tian","year":"2012","unstructured":"Tian S, Wang J, Li Y et al (2012) Drug-likeness analysis of traditional chinese medicines: prediction of drug-likeness using machine learning approaches. Mol Pharm 9:2875\u20132886","journal-title":"Mol Pharm"},{"key":"479_CR5","doi-asserted-by":"publisher","first-page":"716","DOI":"10.1021\/mp400450m","volume":"11","author":"D Li","year":"2014","unstructured":"Li D, Chen L, Li Y et al (2014) ADMET Evaluation in Drug Discovery. 13. Development of in silico prediction models for P-Glycoprotein Substrates. Mol Pharm 11:716\u2013726","journal-title":"Mol Pharm"},{"key":"479_CR6","first-page":"468","volume":"4","author":"JB Mitchell","year":"2014","unstructured":"Mitchell JB (2014) Machine learning methods in chemoinformatics, Wiley Interdisciplinary Reviews: computational Molecular. Science 4:468\u2013481","journal-title":"Science"},{"key":"479_CR7","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1177\/1087057116657513","volume":"22","author":"F Montanari","year":"2017","unstructured":"Montanari F, Cseke A, Wlcek K et al (2017) Virtual screening of drugbank reveals two drugs as new BCRP inhibitors. Slas Discov 22:86\u201393","journal-title":"Slas Discov"},{"key":"479_CR8","doi-asserted-by":"publisher","first-page":"721","DOI":"10.1080\/1062936X.2016.1229691","volume":"27","author":"Y Ren","year":"2016","unstructured":"Ren Y, Zhou L, Yang L et al (2016) Predicting the aquatic toxicity mode of action using logistic regression and linear discriminant analysis. SAR QSAR Environ Res 27:721\u2013746","journal-title":"SAR QSAR Environ Res"},{"key":"479_CR9","doi-asserted-by":"publisher","first-page":"2048","DOI":"10.1021\/ci0340916","volume":"43","author":"VV Zernov","year":"2003","unstructured":"Zernov VV, Balakin K, Ivaschenko AA et al (2003) Drug discovery using support vector machines. The case studies of drug-likeness, agrochemical-likeness, and enzyme inhibition predictions. J Chem Inform Comput Sci 43:2048\u20132056","journal-title":"J Chem Inform Comput Sci"},{"key":"479_CR10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1021\/ci050520j","volume":"47","author":"Q-Y Zhang","year":"2007","unstructured":"Zhang Q-Y, Aires-de-Sousa J (2007) Random forest prediction of mutagenicity from empirical physicochemical descriptors. J Chem Inf Model 47:1\u20138","journal-title":"J Chem Inf Model"},{"key":"479_CR11","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1186\/s13321-016-0151-5","volume":"8","author":"J Alvarsson","year":"2016","unstructured":"Alvarsson J, Lampa S, Schaal W et al (2016) Large-scale ligand-based predictive modelling using support vector machines. J Cheminform 8:39","journal-title":"J Cheminform"},{"key":"479_CR12","doi-asserted-by":"publisher","first-page":"2408","DOI":"10.1021\/ci7002076","volume":"47","author":"T Hou","year":"2007","unstructured":"Hou T, Wang J, Li Y (2007) ADME evaluation in drug discovery 8 The prediction of human intestinal absorption by a support vector machine. J Chem Inform Model 47:2408\u20132415","journal-title":"J Chem Inform Model"},{"key":"479_CR13","doi-asserted-by":"publisher","first-page":"2840","DOI":"10.1021\/ci300400a","volume":"52","author":"C Xu","year":"2012","unstructured":"Xu C, Cheng F, Chen L et al (2012) In silico prediction of chemical ames mutagenicity. J Chem Inf Model 52:2840\u20132847","journal-title":"J Chem Inf Model"},{"key":"479_CR14","doi-asserted-by":"publisher","first-page":"2262","DOI":"10.1021\/jm050200r","volume":"49","author":"F Lombardo","year":"2006","unstructured":"Lombardo F, Obach RS, DiCapua FM et al (2006) A hybrid mixture discriminant analysis-random forest computational model for the prediction of volume of distribution of drugs in human. J Med Chem 49:2262\u20132267","journal-title":"J Med Chem"},{"key":"479_CR15","doi-asserted-by":"publisher","first-page":"1947","DOI":"10.1021\/ci034160g","volume":"43","author":"V Svetnik","year":"2003","unstructured":"Svetnik V, Liaw A, Tong C et al (2003) Random forest: a classification and regression tool for compound classification and QSAR modeling. J Chem Inf Comput Sci 43:1947\u20131958","journal-title":"J Chem Inf Comput Sci"},{"key":"479_CR16","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1002\/minf.201501008","volume":"35","author":"E Gawehn","year":"2016","unstructured":"Gawehn E, Hiss JA, Schneider G (2016) Deep learning in drug discovery. Mol Inform 35:3\u201314","journal-title":"Mol Inform"},{"key":"479_CR17","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, Engkvist O, Wang Y et al (2018) The rise of deep learning in drug discovery. Drug Discov Today 23:1241\u20131250","journal-title":"Drug Discov Today"},{"key":"479_CR18","unstructured":"Dahl GE, Jaitly N, Salakhutdinov R. Multi-task Neural Networks for QSAR Predictions, arXiv preprints arXiv:1406.1231 2014"},{"issue":"3","key":"479_CR19","doi-asserted-by":"publisher","first-page":"919","DOI":"10.1093\/bib\/bbz042","volume":"21","author":"M Sun","year":"2019","unstructured":"Sun M, Zhao S, Gilvary C et al (2019) Graph convolutional networks for computational drug development and discovery. Brief Bioinform 21(3):919\u2013935","journal-title":"Brief Bioinform"},{"key":"479_CR20","doi-asserted-by":"publisher","first-page":"418","DOI":"10.1038\/s41563-019-0332-5","volume":"18","author":"B Bhhatarai","year":"2019","unstructured":"Bhhatarai B, Walters WP, Hop C et al (2019) Opportunities and challenges using artificial intelligence in ADME\/Tox. Nat Mater 18:418\u2013422","journal-title":"Nat Mater"},{"key":"479_CR21","doi-asserted-by":"publisher","first-page":"1680","DOI":"10.1016\/j.drudis.2017.08.010","volume":"22","author":"L Zhang","year":"2017","unstructured":"Zhang L, Tan JJ, Han D et al (2017) From machine learning to deep learning: progress in machine intelligence for rational drug discovery. Drug Discovery Today 22:1680\u20131685","journal-title":"Drug Discovery Today"},{"key":"479_CR22","doi-asserted-by":"publisher","first-page":"4336","DOI":"10.1021\/acs.molpharmaceut.8b00110","volume":"15","author":"X Li","year":"2018","unstructured":"Li X, Xu Y, Lai L et al (2018) Prediction of human cytochrome P450 inhibition using a multitask deep autoencoder neural network. Mol Pharm 15:4336\u20134345","journal-title":"Mol Pharm"},{"key":"479_CR23","doi-asserted-by":"publisher","first-page":"2085","DOI":"10.1021\/acs.jcim.5b00238","volume":"55","author":"Y Xu","year":"2015","unstructured":"Xu Y, Dai Z, Chen F et al (2015) Deep learning for drug-induced liver injury. J Chem Inf Model 55:2085\u20132093","journal-title":"J Chem Inf Model"},{"key":"479_CR24","doi-asserted-by":"publisher","first-page":"3370","DOI":"10.1021\/acs.jcim.9b00237","volume":"59","author":"K Yang","year":"2019","unstructured":"Yang K, Swanson K, Jin W et al (2019) Analyzing learned molecular representations for property prediction. J Chem Inf Model 59:3370\u20133388","journal-title":"J Chem Inf Model"},{"key":"479_CR25","doi-asserted-by":"publisher","first-page":"263","DOI":"10.1021\/ci500747n","volume":"55","author":"J Ma","year":"2015","unstructured":"Ma J, Sheridan RP, Liaw A et al (2015) Deep neural nets as a method for quantitative structure-activity relationships. J Chem Inf Model 55:263\u2013274","journal-title":"J Chem Inf Model"},{"key":"479_CR26","unstructured":"Flam-Shepherd D, Wu T, Friederich P et al. Neural Message Passing on High Order Paths, arXiv preprint arXiv:2002.10413"},{"key":"479_CR27","doi-asserted-by":"crossref","unstructured":"Xiong Z, Wang D, Liu X et al. Pushing the boundaries of molecular representation for drug discovery with the graph attention mechanism. J Med Chem. 2019.","DOI":"10.1021\/acs.jmedchem.9b00959"},{"key":"479_CR28","unstructured":"Chen P, Liu W, Hsieh C-Y et al. Utilizing Edge Features in Graph Neural Networks via Variational Information Maximization, arXiv preprint arXiv:.05488 2019"},{"key":"479_CR29","doi-asserted-by":"crossref","unstructured":"Chen B, Barzilay R, Jaakkola T. Path-Augmented Graph Transformer Network, arXiv preprint arXiv:.12712 2019","DOI":"10.26434\/chemrxiv.8214422"},{"key":"479_CR30","unstructured":"Shang C, Liu Q, Chen K-S et al. Edge attention-based multi-relational graph convolutional networks, arXiv preprint arXiv:.04944 2018"},{"key":"479_CR31","unstructured":"Li J, Cai D, He X. Learning graph-level representation for drug discovery, arXiv preprint arXiv:.03741 2017"},{"key":"479_CR32","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1039\/C7SC02664A","volume":"9","author":"Z Wu","year":"2018","unstructured":"Wu Z, Ramsundar B, Feinberg EN et al (2018) MoleculeNet: a benchmark for molecular machine learning. Chem Sci 9:513\u2013530","journal-title":"Chem Sci"},{"key":"479_CR33","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1021\/acs.jcim.9b00587","volume":"60","author":"V Korolev","year":"2020","unstructured":"Korolev V, Mitrofanov AA, Korotcov A et al (2020) Graph convolutional neural networks as\u201d general-purpose\u201d property predictors: the universality and limits of applicability. J Chem Inf Model 60:22\u201328","journal-title":"J Chem Inf Model"},{"key":"479_CR34","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-019-0407-y","volume":"12","author":"M Withnall","year":"2020","unstructured":"Withnall M, Lindel\u00f6f E, Engkvist O et al (2020) Building attention and edge message passing neural networks for bioactivity and physical\u2013chemical property prediction. J Cheminform 12:1","journal-title":"J Cheminform"},{"key":"479_CR35","doi-asserted-by":"publisher","first-page":"4371","DOI":"10.1021\/acs.molpharmaceut.7b01144","volume":"15","author":"P Hop","year":"2018","unstructured":"Hop P, Allgood B, Yu J (2018) Geometric deep learning autonomously learns chemical features that outperform those engineered by domain experts. Mol Pharm 15:4371\u20134377","journal-title":"Mol Pharm"},{"key":"479_CR36","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 et al (2018) Large-scale comparison of machine learning methods for drug target prediction on ChEMBL. Chem Sci 9:5441\u20135451","journal-title":"Chem Sci"},{"key":"479_CR37","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-020-0414-z","volume":"12","author":"B Tang","year":"2020","unstructured":"Tang B, Kramer ST, Fang M et al (2020) A self-attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility. J Cheminform 12:1\u20139","journal-title":"J Cheminform"},{"key":"479_CR38","doi-asserted-by":"publisher","first-page":"889","DOI":"10.1021\/mp100465q","volume":"8","author":"L Chen","year":"2011","unstructured":"Chen L, Li Y, Zhao Q et al (2011) ADME evaluation in drug discovery 10 predictions of P-glycoprotein inhibitors using recursive partitioning and naive bayesian classification techniques. Mol Pharm 8:889\u2013900","journal-title":"Mol Pharm"},{"key":"479_CR39","doi-asserted-by":"publisher","first-page":"4587","DOI":"10.1021\/acs.jcim.9b00801","volume":"59","author":"Z Wu","year":"2019","unstructured":"Wu Z, Lei T, Shen C et al (2019) ADMET evaluation in drug discovery. 19. Reliable prediction of human cytochrome P450 inhibition using artificial intelligence approaches. J Chem Inform Model 59:4587\u20134601","journal-title":"J Chem Inform Model"},{"key":"479_CR40","doi-asserted-by":"publisher","first-page":"2353","DOI":"10.1021\/acs.jcim.6b00591","volume":"56","author":"RP Sheridan","year":"2016","unstructured":"Sheridan RP, Wang WM, Liaw A et al (2016) Extreme gradient boosting as a method for quantitative structure\u2013activity relationships. J Chem Inf Model 56:2353\u20132360","journal-title":"J Chem Inf Model"},{"key":"479_CR41","doi-asserted-by":"publisher","first-page":"4150","DOI":"10.1021\/acs.jcim.9b00633","volume":"59","author":"J Zhang","year":"2019","unstructured":"Zhang J, Mucs D, Norinder U et al (2019) LightGBM: an effective and scalable algorithm for prediction of chemical toxicity-application to the TOX21 and mutagenicity data sets. J Chem Inf Model 59:4150\u20134158","journal-title":"J Chem Inf Model"},{"key":"479_CR42","doi-asserted-by":"publisher","first-page":"3935","DOI":"10.1021\/acs.molpharmaceut.7b00631","volume":"14","author":"T Lei","year":"2017","unstructured":"Lei T, Sun H, Kang Y et al (2017) ADMET evaluation in drug discovery. 18. Reliable prediction of chemical-induced urinary tract toxicity by boosting machine learning-approaches. Mol Pharm 14:3935\u20133953","journal-title":"Mol Pharm"},{"key":"479_CR43","unstructured":"Landrum G. RDKit: Open-source cheminformatics. 2013"},{"key":"479_CR44","doi-asserted-by":"publisher","first-page":"1466","DOI":"10.1002\/jcc.21707","volume":"32","author":"CW Yap","year":"2011","unstructured":"Yap CW (2011) PaDEL-descriptor: an open source software to calculate molecular descriptors and fingerprints. J Comput Chem 32:1466\u20131474","journal-title":"J Comput Chem"},{"key":"479_CR45","doi-asserted-by":"publisher","first-page":"S69","DOI":"10.1016\/j.toxlet.2017.07.175","volume":"280","author":"G Klambauer","year":"2017","unstructured":"Klambauer G, Unterthiner T, Mayr A et al (2017) DeepTox: toxicity prediction using deep learning. Toxicol Lett 280:S69\u2013S69","journal-title":"Toxicol Lett"},{"key":"479_CR46","doi-asserted-by":"publisher","first-page":"595","DOI":"10.1007\/s10822-016-9938-8","volume":"30","author":"S Kearnes","year":"2016","unstructured":"Kearnes S, McCloskey K, Berndl M et al (2016) Molecular graph convolutions: moving beyond fingerprints. J Comput Aided Mol Des 30:595\u2013608","journal-title":"J Comput Aided Mol Des"},{"key":"479_CR47","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1186\/s13321-017-0226-y","volume":"9","author":"A Koutsoukas","year":"2017","unstructured":"Koutsoukas A, Monaghan KJ, Li X et al (2017) Deep-learning: investigating deep neural networks hyper-parameters and comparison of performance to shallow methods for modeling bioactivity data. J Cheminform 9:42","journal-title":"J Cheminform"},{"key":"479_CR48","unstructured":"Zeiler MD. Adadelta: an adaptive learning rate method, arXiv preprint arXiv:1212.5701 2012"},{"key":"479_CR49","doi-asserted-by":"publisher","first-page":"1882","DOI":"10.1021\/ci0341161","volume":"43","author":"E Byvatov","year":"2003","unstructured":"Byvatov E, Fechner U, Sadowski J et al (2003) Comparison of support vector machine and artificial neural network systems for drug\/nondrug classification. J Chem Inf Comput Sci 43:1882\u20131889","journal-title":"J Chem Inf Comput Sci"},{"key":"479_CR50","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1002\/1521-3838(200110)20:3<227::AID-QSAR227>3.0.CO;2-Y","volume":"20","author":"R Czerminski","year":"2001","unstructured":"Czerminski R, Yasri A, Hartsough D (2001) Use of support vector Machine in pattern classification: application to QSAR studies. Quant Struct-Act Relat 20:227\u2013240","journal-title":"Quant Struct-Act Relat"},{"key":"479_CR51","doi-asserted-by":"crossref","unstructured":"Chen T, Guestrin C. Xgboost: A scalable tree boosting system. In: Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016, p. 785-794. ACM","DOI":"10.1145\/2939672.2939785"},{"key":"479_CR52","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-020-00421-y","volume":"12","author":"D Jiang","year":"2020","unstructured":"Jiang D, Lei T, Wang Z et al (2020) ADMET evaluation in drug discovery 20 Prediction of breast cancer resistance protein inhibition through machine learning. J Cheminform 12:1","journal-title":"J Cheminform"},{"key":"479_CR53","doi-asserted-by":"publisher","first-page":"3714","DOI":"10.1021\/acs.jcim.9b00541","volume":"59","author":"Z-Y Yang","year":"2019","unstructured":"Yang Z-Y, Yang Z-J, Dong J et al (2019) Structural analysis and identification of colloidal aggregators in drug discovery. J Chem Inf Model 59:3714\u20133726","journal-title":"J Chem Inf Model"},{"key":"479_CR54","unstructured":"Gilmer J, Schoenholz SS, Riley PF et al. Neural message passing for quantum chemistry. In: Proceedings of the 34th International Conference on Machine Learning-Volume 70. 2017, p. 1263-1272. JMLR. org"},{"key":"479_CR55","unstructured":"Kipf TN, Welling M. Semi-supervised classification with graph convolutional networks, arXiv preprint arXiv:1609.02907 2016"},{"key":"479_CR56","unstructured":"Veli\u010dkovi\u0107 P, Cucurull G, Casanova A et al. Graph attention networks, arXiv preprint arXiv:1710.10903. 2017."},{"key":"479_CR57","doi-asserted-by":"publisher","first-page":"064101","DOI":"10.1063\/1.5058172","volume":"5","author":"Y Li","year":"2018","unstructured":"Li Y, Cong Y, Feng G et al (2018) The impact of interior dielectric constant and entropic change on HIV-1 complex binding free energy prediction. Struct Dyn. 5:064101","journal-title":"Struct Dyn."},{"key":"479_CR58","first-page":"2825","volume":"12","author":"F Pedregosa","year":"2011","unstructured":"Pedregosa F, Varoquaux G, Gramfort A et al (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825\u20132830","journal-title":"J Mach Learn Res"},{"key":"479_CR59","unstructured":"Wang M, Yu L, Zheng D et al. Deep graph library: towards efficient and scalable deep learning on graphs, arXiv preprint arXiv:1909.01315 2019"},{"key":"479_CR60","doi-asserted-by":"crossref","unstructured":"Bergstra J, Yamins D, Cox DD. Hyperopt: A python library for optimizing the hyperparameters of machine learning algorithms. In: Proceedings of the 12th Python in science conference. 2013, p. 13-20. Citeseer","DOI":"10.25080\/Majora-8b375195-003"},{"key":"479_CR61","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1186\/s13321-017-0232-0","volume":"9","author":"EB Lenselink","year":"2017","unstructured":"Lenselink EB, ten Dijke N, Bongers B et al (2017) Beyond the hype: deep neural networks outperform established methods using a ChEMBL bioactivity benchmark set. J Cheminform 9:45","journal-title":"J Cheminform"},{"key":"479_CR62","doi-asserted-by":"publisher","first-page":"647","DOI":"10.4155\/fmc.11.23","volume":"3","author":"U Johansson","year":"2011","unstructured":"Johansson U, S\u00f6nstr\u00f6d C, Norinder U et al (2011) Trade-off between accuracy and interpretability for predictive in silico modeling. Future Med Chem 3:647\u2013663","journal-title":"Future Med Chem"},{"key":"479_CR63","doi-asserted-by":"publisher","first-page":"269","DOI":"10.1021\/acs.jcim.5b00229","volume":"56","author":"T Fujita","year":"2016","unstructured":"Fujita T, Winkler DA (2016) Understanding the roles of the \u201ctwo QSARs\u201d. J Chem Inf Model 56:269\u2013274","journal-title":"J Chem Inf Model"},{"key":"479_CR64","doi-asserted-by":"publisher","first-page":"2618","DOI":"10.1021\/acs.jcim.7b00274","volume":"57","author":"P Polishchuk","year":"2017","unstructured":"Polishchuk P (2017) Interpretation of quantitative structure-activity relationship models: past, present, and future. J Chem Inform Model 57:2618\u20132639","journal-title":"J Chem Inform Model"},{"key":"479_CR65","doi-asserted-by":"crossref","unstructured":"Rodr\u00edguez-P\u00e9rez R, Bajorath J. Interpretation of compound activity predictions from complex machine learning models using local approximations and Shapley values. J Med Chem. 2019.","DOI":"10.1021\/acs.jmedchem.9b01101"},{"key":"479_CR66","unstructured":"Lundberg SM, Lee S-I. A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems. 2017, p. 4765-4774"},{"key":"479_CR67","unstructured":"Xiong E. OpenDrugAI\/AttentiveFP. https:\/\/github.com\/OpenDrugAI\/AttentiveFP\/blob\/master\/code\/2_Physiology_or_Toxicity_ToxCast.ipynb. Accessed 8 July 2020."},{"key":"479_CR68","doi-asserted-by":"publisher","first-page":"1253","DOI":"10.1021\/acs.jcim.8b00785","volume":"59","author":"J Wenzel","year":"2019","unstructured":"Wenzel J, Matter H, Schmidt F (2019) Predictive multitask deep neural network models for ADME-Tox properties: learning from large data sets. J Chem Inf Model 59:1253\u20131268","journal-title":"J Chem Inf Model"},{"key":"479_CR69","doi-asserted-by":"publisher","first-page":"1062","DOI":"10.1021\/acs.jcim.8b00685","volume":"59","author":"S Sosnin","year":"2018","unstructured":"Sosnin S, Karlov D, Tetko IV et al (2018) Comparative study of multitask toxicity modeling on a broad chemical space. J Chem Inf Model 59:1062\u20131072","journal-title":"J Chem Inf Model"},{"key":"479_CR70","doi-asserted-by":"crossref","unstructured":"Chuang KV, Gunsalus LM, Keiser MJ. Learning Molecular Representations for Medicinal Chemistry, J Med Chem 2020","DOI":"10.1021\/acs.jmedchem.0c00385"},{"key":"479_CR71","doi-asserted-by":"publisher","first-page":"1000","DOI":"10.1021\/ci034243x","volume":"44","author":"JS Delaney","year":"2004","unstructured":"Delaney JS (2004) ESOL: estimating aqueous solubility directly from molecular structure. J Chem Inf Comput Sci 44:1000\u20131005","journal-title":"J Chem Inf Comput Sci"},{"key":"479_CR72","doi-asserted-by":"crossref","unstructured":"Li R, Wang S, Zhu F et al. Adaptive graph convolutional neural networks, arXiv preprint arXiv:1801.03226 2018","DOI":"10.1609\/aaai.v32i1.11691"}],"container-title":["Journal of Cheminformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13321-020-00479-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13321-020-00479-8\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13321-020-00479-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,29]],"date-time":"2023-01-29T05:07:01Z","timestamp":1674968821000},"score":1,"resource":{"primary":{"URL":"https:\/\/jcheminf.biomedcentral.com\/articles\/10.1186\/s13321-020-00479-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,2,17]]},"references-count":72,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2021,12]]}},"alternative-id":["479"],"URL":"https:\/\/doi.org\/10.1186\/s13321-020-00479-8","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-79416\/v1","asserted-by":"object"},{"id-type":"doi","id":"10.21203\/rs.3.rs-81439\/v1","asserted-by":"object"}]},"ISSN":["1758-2946"],"issn-type":[{"value":"1758-2946","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,2,17]]},"assertion":[{"value":"21 September 2020","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"26 November 2020","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 February 2021","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare that they have no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"12"}}