{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T23:41:33Z","timestamp":1782517293442,"version":"3.54.5"},"reference-count":52,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2022,12,19]],"date-time":"2022-12-19T00:00:00Z","timestamp":1671408000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"name":"Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism","award":["2021 Sci & Tech 03-28"],"award-info":[{"award-number":["2021 Sci & Tech 03-28"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["82173746"],"award-info":[{"award-number":["82173746"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["81872800"],"award-info":[{"award-number":["81872800"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2019YFA0904800"],"award-info":[{"award-number":["2019YFA0904800"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,1,19]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Qualitative or quantitative prediction models of structure\u2013activity relationships based on graph neural networks (GNNs) are prevalent in drug discovery applications and commonly have excellently predictive power. However, the network information flows of GNNs are highly complex and accompanied by poor interpretability. Unfortunately, there are relatively less studies on GNN attributions, and their developments in drug research are still at the early stages. In this work, we adopted several advanced attribution techniques for different GNN frameworks and applied them to explain multiple drug molecule property prediction tasks, enabling the identification and visualization of vital chemical information in the networks. Additionally, we evaluated them quantitatively with attribution metrics such as accuracy, sparsity, fidelity and infidelity, stability and sensitivity; discussed their applicability and limitations; and provided an open-source benchmark platform for researchers. The results showed that all attribution techniques were effective, while those directly related to the predicted labels, such as integrated gradient, preferred to have better attribution performance. These attribution techniques we have implemented could be directly used for the vast majority of chemical GNN interpretation tasks.<\/jats:p>","DOI":"10.1093\/bib\/bbac577","type":"journal-article","created":{"date-parts":[[2022,12,20]],"date-time":"2022-12-20T07:52:01Z","timestamp":1671522721000},"source":"Crossref","is-referenced-by-count":14,"title":["Identification of vital chemical information via visualization of graph neural networks"],"prefix":"10.1093","volume":"24","author":[{"given":"Yimeng","family":"Wang","sequence":"first","affiliation":[{"name":"School of Pharmacy, East China University of Science and Technology Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, , 130 Meilong Road, Shanghai 200237 , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengting","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Pharmacy, East China University of Science and Technology Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, , 130 Meilong Road, Shanghai 200237 , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hua","family":"Deng","sequence":"additional","affiliation":[{"name":"School of Pharmacy, East China University of Science and Technology Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, , 130 Meilong Road, Shanghai 200237 , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weihua","family":"Li","sequence":"additional","affiliation":[{"name":"School of Pharmacy, East China University of Science and Technology Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, , 130 Meilong Road, Shanghai 200237 , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3095-6006","authenticated-orcid":false,"given":"Zengrui","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Pharmacy, East China University of Science and Technology Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, , 130 Meilong Road, Shanghai 200237 , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2340-1109","authenticated-orcid":false,"given":"Yun","family":"Tang","sequence":"additional","affiliation":[{"name":"School of Pharmacy Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, , East China University of Science and Technology, 130 Meilong Road, Shanghai 200237 , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9648-844X","authenticated-orcid":false,"given":"Guixia","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Pharmacy, East China University of Science and Technology Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, , 130 Meilong Road, Shanghai 200237 , China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2022,12,19]]},"reference":[{"key":"2023011917121408100_ref1","doi-asserted-by":"crossref","first-page":"bbab430","DOI":"10.1093\/bib\/bbab430","article-title":"Artificial intelligence in drug discovery: applications and techniques","volume":"23","author":"Deng","year":"2022","journal-title":"Brief Bioinform"},{"key":"2023011917121408100_ref2","doi-asserted-by":"crossref","first-page":"1538","DOI":"10.1016\/j.drudis.2018.05.010","article-title":"Machine learning in chemoinformatics and drug discovery","volume":"23","author":"Lo","year":"2018","journal-title":"Drug Discov Today"},{"key":"2023011917121408100_ref3","doi-asserted-by":"crossref","first-page":"bbaa321","DOI":"10.1093\/bib\/bbaa321","article-title":"Do we need different machine learning algorithms for QSAR modeling? A comprehensive assessment of 16 machine learning algorithms on 14 QSAR data sets","volume":"22","author":"Wu","year":"2021","journal-title":"Brief Bioinform"},{"key":"2023011917121408100_ref4","doi-asserted-by":"crossref","first-page":"780","DOI":"10.1002\/cpt.1795","article-title":"Will artificial intelligence for drug discovery impact clinical pharmacology?","volume":"107","author":"Zhavoronkov","year":"2020","journal-title":"Clin Pharmacol Therap"},{"key":"2023011917121408100_ref5","article-title":"General Data Protection Regulation (GDPR)","journal-title":"General data protection regulation (GDPR) \u2013 official legal text"},{"key":"2023011917121408100_ref6","first-page":"5898","article-title":"Evaluating attribution for graph neural networks","volume":"33","author":"Sanchez-Lengeling","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"2023011917121408100_ref7","doi-asserted-by":"crossref","first-page":"573","DOI":"10.1038\/s42256-020-00236-4","article-title":"Drug discovery with explainable artificial intelligence","volume":"2","author":"Jim\u00e9nez-Luna","year":"2020","journal-title":"Nat Mach Intell"},{"key":"2023011917121408100_ref8","first-page":"1","volume-title":"Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems","author":"Liao","year":"2020"},{"key":"2023011917121408100_ref9","doi-asserted-by":"crossref","first-page":"331","DOI":"10.1007\/978-3-030-28954-6_18","volume-title":"Explainable AI: Interpreting, Explaining and Visualizing Deep Learning","author":"Preuer","year":"2019"},{"key":"2023011917121408100_ref10","doi-asserted-by":"crossref","first-page":"1324","DOI":"10.1021\/acs.jcim.8b00825","article-title":"Interpretation of QSAR models by coloring atoms according to changes in predicted activity: how robust is it?","volume":"59","author":"Sheridan","year":"2019","journal-title":"J Chem Inf Model"},{"key":"2023011917121408100_ref11","doi-asserted-by":"crossref","first-page":"1083","DOI":"10.1021\/acs.jcim.0c01344","article-title":"Coloring molecules with explainable artificial intelligence for preclinical relevance assessment","volume":"61","author":"Jim\u00e9nez-Luna","year":"2021","journal-title":"J Chem Inf Model"},{"key":"2023011917121408100_ref12","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1021\/acs.jcim.1c01263","article-title":"Interpretation of structure\u2013activity relationships in real-world drug design data sets using explainable artificial intelligence","volume":"62","author":"Harren","year":"2022","journal-title":"J Chem Inf Model"},{"key":"2023011917121408100_ref13","volume-title":"RDKit: Open-Source Cheminformatics Software"},{"key":"2023011917121408100_ref14","volume-title":"MolVS: molecule validation and standardization"},{"key":"2023011917121408100_ref15","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","year":"2020","journal-title":"J Med Chem"},{"key":"2023011917121408100_ref16","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1016\/S0169-409X(96)00423-1","article-title":"Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings","volume":"23","author":"Lipinski","year":"1997","journal-title":"Adv Drug Deliv Rev"},{"key":"2023011917121408100_ref17","doi-asserted-by":"crossref","first-page":"1189","DOI":"10.1021\/jm950902o","article-title":"Exploring QSAR: hydrophobic, electronic, and steric constants C. Hansch, A. Leo, and D. Hoekman. American Chemical Society, Washington, DC. 1995. Xix + 348 pp. 22 \u00d7 28.5 cm. Exploring QSAR: fundamentals and applications in chemistry and biology. C. Hansch and A. Leo. American Chemical Society, Washington, DC. 1995. Xvii + 557 pp. 18.5 \u00d7 26 cm. ISBN 0-8412-2993-7 (set). $99.95 (set)","volume":"39","author":"Martin","year":"1996","journal-title":"J Med Chem"},{"key":"2023011917121408100_ref18","doi-asserted-by":"crossref","first-page":"2140","DOI":"10.1021\/ci700257y","article-title":"Computation of octanol\u2212water partition coefficients by guiding an additive model with knowledge","volume":"47","author":"Cheng","year":"2007","journal-title":"J Chem Inf Model"},{"key":"2023011917121408100_ref19","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1016\/j.ejps.2012.10.019","article-title":"Large, chemically diverse dataset of logP measurements for benchmarking studies","volume":"48","author":"Martel","year":"2013","journal-title":"Eur J Pharm Sci"},{"key":"2023011917121408100_ref20","doi-asserted-by":"crossref","first-page":"937","DOI":"10.1007\/s10822-018-0170-6","article-title":"Overview of the SAMPL6 host\u2013guest binding affinity prediction challenge","volume":"32","author":"Rizzi","year":"2018","journal-title":"J Comput Aided Mol Des"},{"key":"2023011917121408100_ref21","doi-asserted-by":"crossref","first-page":"868","DOI":"10.1021\/ci990307l","article-title":"Prediction of physicochemical parameters by atomic contributions","volume":"39","author":"Wildman","year":"1999","journal-title":"J Chem Inf Comput Sci"},{"key":"2023011917121408100_ref22","article-title":"European Parliament and Council of the European Union, Directive 2006\/25\/EU of the European Parliament and of the Council of 5 April 2006 on the minimum health and safety requirements regarding the exposure of workers to risks arising from physical agents (artificial optical radiation)","volume-title":"Official Journal of the European Union L","author":""},{"key":"2023011917121408100_ref23","doi-asserted-by":"crossref","first-page":"1061","DOI":"10.1021\/ci5000467","article-title":"In silico prediction of chemical acute oral toxicity using multi-classification methods","volume":"54","author":"Li","year":"2014","journal-title":"J Chem Inf Model"},{"key":"2023011917121408100_ref24","doi-asserted-by":"crossref","first-page":"1913","DOI":"10.1021\/tx900189p","article-title":"Quantitative structure\u2212activity relationship modeling of rat acute toxicity by oral exposure","volume":"22","author":"Zhu","year":"2009","journal-title":"Chem Res Toxicol"},{"key":"2023011917121408100_ref25","doi-asserted-by":"crossref","first-page":"123","DOI":"10.18097\/PBMC20196502123","article-title":"QSAR investigation of acute toxicity of organic compounds during oral administration to mice","volume":"65","author":"Tinkov","year":"2019","journal-title":"Biomed Khim"},{"key":"2023011917121408100_ref26","doi-asserted-by":"crossref","first-page":"W53","DOI":"10.1093\/nar\/gku401","article-title":"ProTox: a web server for the in silico prediction of rodent oral toxicity","volume":"42","author":"Drwal","year":"2014","journal-title":"Nucleic Acids Res"},{"key":"2023011917121408100_ref27","doi-asserted-by":"crossref","first-page":"2310","DOI":"10.1021\/ci300245q","article-title":"ToxAlerts: a web server of structural alerts for toxic chemicals and compounds with potential adverse reactions","volume":"52","author":"Sushko","year":"2012","journal-title":"J Chem Inf Model"},{"key":"2023011917121408100_ref28","article-title":"Semi-supervised classification with graph convolutional networks","volume-title":"arXiv","author":""},{"key":"2023011917121408100_ref29","article-title":"Inductive representation learning on large graphs","author":"","journal-title":"31st Conference on Neural Information Processing Systems"},{"key":"2023011917121408100_ref30","first-page":"20","article-title":"Graph attention networks","volume":"1050","author":"Velickovic","year":"2017","journal-title":"Stat"},{"key":"2023011917121408100_ref31","article-title":"Empirical evaluation of gated recurrent neural networks on sequence modeling","author":"Chung","year":"2014","journal-title":"arXiv"},{"key":"2023011917121408100_ref32","article-title":"Hyperparameters optimization in deep convolutional neural network\/bayesian approach with gaussian process prior","author":"Murugan","year":"2017","journal-title":"arXiv"},{"key":"2023011917121408100_ref33","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2014","journal-title":"arXiv"},{"key":"2023011917121408100_ref34","article-title":"Deep graph library: a graph-centric, highly-performant package for graph neural networks","author":"Wang","year":"2019","journal-title":"arXiv"},{"key":"2023011917121408100_ref35","doi-asserted-by":"crossref","first-page":"27233","DOI":"10.1021\/acsomega.1c04017","article-title":"DGL-LifeSci: an open-source toolkit for deep learning on graphs in life science","volume":"6","author":"Li","year":"2021","journal-title":"ACS Omega"},{"key":"2023011917121408100_ref36","article-title":"Pytorch: an imperative style, high-performance deep learning library","author":"","journal-title":"33rd Conference on Neural Information Processing Systems"},{"key":"2023011917121408100_ref37","article-title":"Explainability in graph neural networks: a taxonomic survey","author":"Yuan","year":"2020","journal-title":"arXiv"},{"key":"2023011917121408100_ref38","first-page":"3319","volume-title":"Proceedings of the 34th International Conference on Machine Learning","author":"Sundararajan","year":"2017"},{"key":"2023011917121408100_ref39","article-title":"GNN explainer: generating explanations for graph neural networks","author":"Ying","journal-title":"33rd Conference on Neural Information Processing Systems"},{"key":"2023011917121408100_ref40","first-page":"12225","article-title":"Pgm-explainer: probabilistic graphical model explanations for graph neural networks","volume":"33","author":"Vu","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"2023011917121408100_ref41","first-page":"376","volume-title":"The 16th international FLAIRS conference","author":"Tsamardinos","year":"2003"},{"key":"2023011917121408100_ref42","article-title":"Bayesian network induction via local neighborhoods","author":"Margaritis","year":"1999","journal-title":"12st Conference on Neural Information Processing Systems"},{"key":"2023011917121408100_ref43","article-title":"The shapley value","volume-title":"Handbook of game theory with economic applications","author":""},{"key":"2023011917121408100_ref44","article-title":"A unified approach to interpreting model predictions","author":"Lundberg","journal-title":"31st Conference on Neural Information Processing Systems"},{"key":"2023011917121408100_ref45","first-page":"12241","volume-title":"Proceedings of the 38th International Conference on Machine Learning","author":"Yuan","year":"2021"},{"key":"2023011917121408100_ref46","first-page":"1","volume-title":"IEEE Transactions on Computational Intelligence and AI in Games","author":"Browne","year":"2012"},{"key":"2023011917121408100_ref47","article-title":"Towards faithfully interpretable nlp systems: How should we define and evaluate faithfulness?","author":"Jacovi","year":"2020","journal-title":"arXiv"},{"key":"2023011917121408100_ref48","doi-asserted-by":"crossref","first-page":"159","DOI":"10.1007\/978-3-319-90403-0_9","volume-title":"Human and Machine Learning: Visible, Explainable, Trustworthy and Transparent","author":"Robnik-\u0160ikonja","year":"2018"},{"key":"2023011917121408100_ref49","first-page":"10772","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Pope","year":"2019"},{"key":"2023011917121408100_ref50","doi-asserted-by":"crossref","first-page":"714","DOI":"10.1093\/bib\/bbq090","article-title":"Principal component analysis based methods in bioinformatics studies","volume":"12","author":"Ma","year":"2011","journal-title":"Brief Bioinform"},{"key":"2023011917121408100_ref51","doi-asserted-by":"crossref","DOI":"10.1021\/acs.jcim.2c00297","article-title":"IDL-PPBopt: a strategy for prediction and optimization of human plasma protein binding of compounds via an interpretable deep learning method","volume":"62","author":"Lou","year":"2022","journal-title":"J Chem Inf Model"},{"key":"2023011917121408100_ref52","first-page":"11624","volume-title":"Proc Natl Acad Sci USA","author":"McCloskey","year":"2019"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/24\/1\/bbac577\/48782800\/bbac577.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/24\/1\/bbac577\/48782800\/bbac577.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,19]],"date-time":"2023-01-19T17:37:21Z","timestamp":1674149841000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbac577\/6936421"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,12,19]]},"references-count":52,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2023,1,19]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbac577","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,1]]},"published":{"date-parts":[[2022,12,19]]},"article-number":"bbac577"}}