{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,30]],"date-time":"2025-07-30T16:30:54Z","timestamp":1753893054529,"version":"3.41.2"},"reference-count":61,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,3,23]],"date-time":"2023-03-23T00:00:00Z","timestamp":1679529600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Accurate prediction of drug response is a crucial step in personalized medicine. Recently, deep learning techniques have been witnessed with significant breakthroughs in a variety of areas including biomedical research and chemogenomic applications. This motivated us to develop a novel deep learning platform to accurately and reliably predict the response of cancer cells to different drug treatments. In the present work, we describe a Java-based implementation of deep neural network method, termed JavaDL, to predict cancer responses to drugs solely based on their chemical features. To this end, we devised a novel cost function and added a regularization term which suppresses overfitting. We also adopted an early stopping strategy to further reduce overfit and improve the accuracy and robustness of our models. To evaluate our method, we compared with several popular machine learning and deep neural network programs and observed that JavaDL either outperformed those methods in model building or obtained comparable predictions. Finally, JavaDL was employed to predict drug responses of several aggressive breast cancer cell lines, and the results showed robust and accurate predictions with <jats:italic>r<\/jats:italic><jats:sup>2<\/jats:sup> as high as 0.81.<\/jats:p>","DOI":"10.3389\/frai.2023.1069353","type":"journal-article","created":{"date-parts":[[2023,3,23]],"date-time":"2023-03-23T07:19:52Z","timestamp":1679555992000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["Development and evaluation of a java-based deep neural network method for drug response predictions"],"prefix":"10.3389","volume":"6","author":[{"given":"Beibei","family":"Huang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lon W. R.","family":"Fong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rajan","family":"Chaudhari","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shuxing","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2023,3,23]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"270","DOI":"10.1016\/j.compbiomed.2017.09.017","article-title":"Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals","volume":"100","author":"Acharya","year":"2018","journal-title":"Comput. Biol.Med."},{"key":"B2","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1038\/s41698-020-0122-1","article-title":"Machine learning approaches to drug response prediction: challenges and recent progress","volume":"4","author":"Adam","year":"2020","journal-title":"NPJ Precis. Oncol."},{"key":"B3","doi-asserted-by":"publisher","first-page":"878","DOI":"10.15252\/msb.20156651","article-title":"Deep learning for computational biology","volume":"12","author":"Angermueller","year":"2016","journal-title":"Mol. Syst. Biol."},{"key":"B4","doi-asserted-by":"publisher","first-page":"879","DOI":"10.1080\/17460441.2017.1353494","article-title":"Representation and identification of activity cliffs","volume":"12","author":"Bajorath","year":"2017","journal-title":"Expert Opin. Drug Discov."},{"key":"B5","first-page":"199","article-title":"Introduction to MOLE DB - on-line molecular descriptors database","volume":"62","author":"Ballabio","year":"2009","journal-title":"Match Commun. Math. Comput. Chemist."},{"key":"B6","doi-asserted-by":"publisher","first-page":"755","DOI":"10.1080\/17460441.2020.1745183","article-title":"The power of deep learning to ligand-based novel drug discovery","volume":"15","author":"Baskin","year":"2020","journal-title":"Expert Opin. Drug Discov"},{"key":"B7","doi-asserted-by":"publisher","first-page":"e1006269","DOI":"10.1371\/journal.pcbi.1006269","article-title":"Evaluating reproducibility of AI algorithms in digital pathology with DAPPER","volume":"15","author":"Bizzego","year":"2019","journal-title":"PLoS Comput. Biol."},{"key":"B8","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1007\/978-1-60327-101-1_3","article-title":"Bayesian regularization of neural networks","volume":"458","author":"Burden","year":"2008","journal-title":"Methods Mol. Biol."},{"key":"B9","first-page":"16","article-title":"Is it a prime time for ai-powered virtual drug screening?","author":"Carpenter","year":"2017","journal-title":"EC Pharmacol. Toxicol"},{"key":"B10","doi-asserted-by":"publisher","first-page":"e0156338","DOI":"10.1371\/journal.pone.0156338","article-title":"Forecasting east asian indices futures via a novel hybrid of wavelet-PCA denoising and artificial neural network models","volume":"11","author":"Chan Phooi M'ng","year":"2016","journal-title":"PLoS ONE"},{"key":"B11","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1080\/17460441.2017.1280024","article-title":"Computational polypharmacology: a new paradigm for drug discovery","volume":"12","author":"Chaudhari","year":"2017","journal-title":"Expert Opin. Drug Discov."},{"key":"B12","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1186\/s12859-018-2534-2","article-title":"Biomedical semantic indexing by deep neural network with multi-task learning","volume":"19","author":"Du","year":"2018","journal-title":"BMC Bioinformatics"},{"key":"B13","doi-asserted-by":"publisher","first-page":"2948","DOI":"10.1021\/ci200271d","article-title":"A critical assessment of combined ligand- and structure-based approaches to HERG channel blocker modeling","volume":"51","author":"Du-Cuny","year":"2011","journal-title":"J. Chem. Inf. Model."},{"key":"B14","doi-asserted-by":"publisher","first-page":"6983","DOI":"10.1016\/j.bmc.2009.08.022","article-title":"Computational modeling of novel inhibitors targeting the Akt pleckstrin homology domain","volume":"17","author":"Du-Cuny","year":"2009","journal-title":"Bioorg. Med. Chem."},{"journal-title":"De novo structure prediction with deep-learning based scoring. In Thirteenth Critical Assessment of Techniques for Protein Structure Prediction","year":"2018","author":"Evans","key":"B15"},{"key":"B16","doi-asserted-by":"publisher","first-page":"S49","DOI":"10.1038\/d41586-019-03846-0","article-title":"Hunting for new drugs with AI","volume":"576","author":"Freedman","year":"2019","journal-title":"Nature"},{"key":"B17","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1002\/minf.201501008","article-title":"Deep learning in drug discovery","volume":"35","author":"Gawehn","year":"2016","journal-title":"Mol. Inform."},{"key":"B18","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1023\/A:1025386326946","article-title":"Rational selection of training and test sets for the development of validated QSAR models","volume":"17","author":"Golbraikh","year":"2003","journal-title":"J. Comput. Aided Mol. Des."},{"key":"B19","doi-asserted-by":"publisher","first-page":"269","DOI":"10.1016\/S1093-3263(01)00123-1","article-title":"Beware of q2!","volume":"20","author":"Golbraikh","year":"2002","journal-title":"J. Mol. Graph. Model."},{"key":"B20","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1038\/s41591-018-0279-0","article-title":"Identifying facial phenotypes of genetic disorders using deep learning","volume":"25","author":"Gurovich","year":"2019","journal-title":"Nat. Med."},{"key":"B21","doi-asserted-by":"publisher","first-page":"11476","DOI":"10.1038\/srep11476","article-title":"Improving prediction of secondary structure, local backbone angles, and solvent accessible surface area of proteins by iterative deep learning","volume":"5","author":"Heffernan","year":"2015","journal-title":"Sci. Rep."},{"key":"B22","doi-asserted-by":"publisher","first-page":"1078","DOI":"10.1111\/cogs.12049","article-title":"Where do features come from?","volume":"38","author":"Hinton","year":"2014","journal-title":"Cogn. Sci."},{"key":"B23","doi-asserted-by":"publisher","first-page":"251","DOI":"10.1016\/0893-6080(91)90009-T","article-title":"Approximation capabilities of multilayer feedforward networks","volume":"4","author":"Hornik","year":"1991","journal-title":"Neural Netw."},{"key":"B24","doi-asserted-by":"publisher","first-page":"1281","DOI":"10.1016\/j.cell.2016.10.039","article-title":"Epigenetic activation of WNT5A drives glioblastoma stem cell differentiation and invasive growth","volume":"167","author":"Hu","year":"2016","journal-title":"Cell."},{"key":"B25","doi-asserted-by":"publisher","first-page":"abm4805","DOI":"10.1126\/science.abm4805","article-title":"Computed structures of core eukaryotic protein complexes","volume":"374","author":"Humphreys","year":"2021","journal-title":"Science."},{"key":"B26","doi-asserted-by":"publisher","first-page":"1062","DOI":"10.1021\/ci500742b","article-title":"Structure-based predictions of activity cliffs","volume":"55","author":"Husby","year":"2015","journal-title":"J. Chem. Inf. Model."},{"key":"B27","doi-asserted-by":"publisher","first-page":"791","DOI":"10.1016\/j.patrec.2019.08.003","article-title":"Multi-projection deep learning network for segmentation of 3D medical images","volume":"125","author":"Indraswari","year":"2019","journal-title":"Pattern Recognit. Lett."},{"key":"B28","doi-asserted-by":"publisher","first-page":"923","DOI":"10.1007\/s10278-018-0099-2","article-title":"Laterality classification of fundus images using interpretable deep neural network","volume":"31","author":"Jang","year":"2018","journal-title":"J. Digit. Imaging"},{"key":"B29","doi-asserted-by":"publisher","first-page":"829","DOI":"10.1007\/s00432-018-02834-7","article-title":"Computer-assisted medical image classification for early diagnosis of oral cancer employing deep learning algorithm","volume":"145","author":"Jeyaraj","year":"2019","journal-title":"J. Cancer Res. Clin. Oncol."},{"key":"B30","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly accurate protein structure prediction with AlphaFold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1245","DOI":"10.1109\/TPAMI.2017.2701831","article-title":"Shakeout: a new approach to regularized deep neural network training","volume":"40","author":"Kang","year":"2018","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B32","doi-asserted-by":"publisher","first-page":"1272","DOI":"10.1111\/cbdd.13188","article-title":"Rational design of isonicotinic acid hydrazide derivatives with antitubercular activity: Machine learning, molecular docking, synthesis and biological testing","volume":"92","author":"Kovalishyn","year":"2018","journal-title":"Chem. Biol. Drug Des."},{"key":"B33","doi-asserted-by":"publisher","first-page":"415","DOI":"10.1186\/s12859-019-3013-0","article-title":"Novel deep learning model for more accurate prediction of drug-drug interaction effects","volume":"20","author":"Lee","year":"2019","journal-title":"BMC Bioinformatics"},{"key":"B34","doi-asserted-by":"publisher","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":"B35","doi-asserted-by":"publisher","first-page":"263","DOI":"10.1021\/ci500747n","article-title":"Deep Neural Nets as a Method for Quantitative Structure-Activity Relationships","volume":"55","author":"Ma","year":"2015","journal-title":"J. Chem. Inf. Model."},{"key":"B36","doi-asserted-by":"publisher","first-page":"450","DOI":"10.1007\/s11684-019-0726-4","article-title":"Survey on deep learning for pulmonary medical imaging","volume":"14","author":"Ma","year":"2019","journal-title":"Front. Med"},{"key":"B37","doi-asserted-by":"publisher","first-page":"1535","DOI":"10.1021\/ci060117s","article-title":"On outliers and activity cliffs - why QSAR often disappoints","volume":"46","author":"Maggiora","year":"2006","journal-title":"J. Chem. Inf. Model."},{"key":"B38","doi-asserted-by":"publisher","first-page":"327","DOI":"10.1038\/nchembio.576","article-title":"An active role for machine learning in drug development","volume":"7","author":"Murphy","year":"2011","journal-title":"Nat. Chem. Biol."},{"key":"B39","doi-asserted-by":"publisher","first-page":"2536","DOI":"10.1021\/ci500092j","article-title":"Curation and analysis of multitargeting agents for polypharmacological modeling","volume":"54","author":"Reddy","year":"2014","journal-title":"J. Chem. Inf. Model."},{"key":"B40","doi-asserted-by":"publisher","first-page":"638","DOI":"10.1038\/s41591-018-0007-9","article-title":"Mechanisms and clinical activity of an EGFR and HER2 exon 20-selective kinase inhibitor in non-small cell lung cancer","volume":"24","author":"Robichaux","year":"2018","journal-title":"Nat. Med."},{"key":"B41","doi-asserted-by":"publisher","first-page":"604","DOI":"10.1038\/nature25978","article-title":"Planning chemical syntheses with deep neural networks and symbolic AI","volume":"555","author":"Segler","year":"2018","journal-title":"Nature"},{"key":"B42","doi-asserted-by":"publisher","first-page":"706","DOI":"10.1038\/s41586-019-1923-7","article-title":"Improved protein structure prediction using potentials from deep learning","volume":"577","author":"Senior","year":"2020","journal-title":"Nature"},{"key":"B43","doi-asserted-by":"publisher","first-page":"604","DOI":"10.1038\/nbt0717-604","article-title":"AI-powered drug discovery captures pharma interest","volume":"35","author":"Smalley","year":"2017","journal-title":"Nat. Biotechnol."},{"key":"B44","doi-asserted-by":"publisher","first-page":"1065","DOI":"10.1021\/acsmedchemlett.8b00437","article-title":"Transforming computational drug discovery with machine learning and AI","volume":"9","author":"Smith","year":"2018","journal-title":"ACS Med. Chem. Lett."},{"key":"B45","unstructured":"Dropout: A simple way to prevent neural networks from overfitting19291958\n            SrivastavaN.\n            HintonG.\n            KrizhevskyA.\n            SutskeverI.\n            SalakhutdinovR.\n          33259321J. Mach. Learn. Res152014"},{"key":"B46","doi-asserted-by":"publisher","first-page":"l5434","DOI":"10.1136\/bmj.l5434","article-title":"New AI laboratory for the NHS","volume":"366","author":"Steventon","year":"2019","journal-title":"BMJ"},{"key":"B47","doi-asserted-by":"publisher","first-page":"1211","DOI":"10.1002\/cmdc.201600067","article-title":"Polypharmacology in drug development: a minireview of current technologies","volume":"11","author":"Tan","year":"","journal-title":"Chem. Med. Chem."},{"key":"B48","doi-asserted-by":"publisher","first-page":"33534","DOI":"10.1038\/srep33534","article-title":"Comprehensive modeling and discovery of mebendazole as a Novel TRAF2- and NCK-interacting Kinase Inhibitor","volume":"6","author":"Tan","year":"","journal-title":"Sci. Rep."},{"key":"B49","doi-asserted-by":"publisher","first-page":"345","DOI":"10.2174\/1389557515666151001154111","article-title":"Past, present, and future of targeting ras for cancer therapies","volume":"16","author":"Tan","year":"2016","journal-title":"Mini Rev. Med. Chem."},{"key":"B50","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1002\/qsar.200390007","article-title":"The importance of being earnest: validation is the absolute essential for successful application and interpretation of QSPR models","volume":"22","author":"Tropsha","year":"2003","journal-title":"Qsar Comb. Sci."},{"key":"B51","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1200\/JCO.2017.35.15_suppl.9062","article-title":"Response of germline and somatic smoothened (SMO) mutations in non-small cell lung cancer (NSCLC) to hedgehog inhibitor vismodegib","volume":"2017","author":"Tsao","year":"2017","journal-title":"J. Clin. Oncol."},{"key":"B52","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1038\/s41573-019-0024-5","article-title":"Applications of machine learning in drug discovery and development","volume":"18","author":"Vamathevan","year":"2019","journal-title":"Nat. Rev. Drug Discovery"},{"key":"B53","doi-asserted-by":"publisher","first-page":"2996","DOI":"10.1021\/acs.jcim.7b00110","article-title":"Theoretical analysis of activity cliffs among benzofuranone-class pim1 inhibitors using the fragment molecular orbital method with molecular mechanics poisson-boltzmann surface area (FMO+MM-PBSA) approach","volume":"57","author":"Watanabe","year":"2017","journal-title":"J. Chem. Inf. Model."},{"key":"B54","doi-asserted-by":"publisher","first-page":"160018","DOI":"10.1002\/minf.201781141","article-title":"Performance of deep and shallow neural networks, the universal approximation theorem, activity cliffs, and QSAR","volume":"36","author":"Winkler","year":"2017","journal-title":"Mol. Inform."},{"key":"B55","doi-asserted-by":"publisher","first-page":"S100","DOI":"10.1038\/d41586-019-02871-3","article-title":"An AI boost for clinical trials","volume":"573","author":"Woo","year":"2019","journal-title":"Nature"},{"key":"B56","doi-asserted-by":"publisher","first-page":"1630","DOI":"10.1021\/ci049869h","article-title":"Effect of molecular descriptor feature selection in support vector machine classification of pharmacokinetic and toxicological properties of chemical agents","volume":"44","author":"Xue","year":"2004","journal-title":"J. Chem. Inf. Comput. Sci."},{"key":"B57","doi-asserted-by":"publisher","first-page":"90","DOI":"10.1016\/j.compbiolchem.2019.03.016","article-title":"Predicting drug-target interaction network using deep learning model","volume":"80","author":"You","year":"2019","journal-title":"Comput. Biol. Chem."},{"key":"B58","doi-asserted-by":"publisher","first-page":"5191","DOI":"10.1093\/bioinformatics\/btz418","article-title":"deepDR: a network-based deep learning approach to in silico drug repositioning","volume":"35","author":"Zeng","year":"2019","journal-title":"Bioinformatics"},{"key":"B59","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1109\/TNNLS.2016.2631481","article-title":"Decoupled ARX and RBF neural network modeling using PCA and GA optimization for nonlinear distributed parameter systems","volume":"29","author":"Zhang","year":"2018","journal-title":"IEEE Trans. Neural Netw. Learn Syst."},{"key":"B60","doi-asserted-by":"publisher","first-page":"1984","DOI":"10.1021\/ci060132x","article-title":"A novel automated lazy learning QSAR (ALL-QSAR) approach: Method development, applications, and virtual screening of chemical databases using validated ALL-QSAR models","volume":"46","author":"Zhang","year":"","journal-title":"J. Chem. Inf. Model."},{"key":"B61","doi-asserted-by":"publisher","first-page":"2713","DOI":"10.1021\/jm050260x","article-title":"Development of quantitative structure-binding affinity relationship models based on novel geometrical chemical descriptors of the protein-ligand interfaces","volume":"49","author":"Zhang","year":"","journal-title":"J. Med. Chem."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2023.1069353\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,3,23]],"date-time":"2023-03-23T07:20:00Z","timestamp":1679556000000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2023.1069353\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,23]]},"references-count":61,"alternative-id":["10.3389\/frai.2023.1069353"],"URL":"https:\/\/doi.org\/10.3389\/frai.2023.1069353","relation":{},"ISSN":["2624-8212"],"issn-type":[{"type":"electronic","value":"2624-8212"}],"subject":[],"published":{"date-parts":[[2023,3,23]]},"article-number":"1069353"}}