Machine learning and deep learning approaches for enhanced prediction of hERG blockade: a comprehensive QSAR modeling study

References

• Craveiro NS, Lopes BS, Tomas L, et al. Drug withdrawal due to safety: a review of the data supporting withdrawal decision. Curr Drug Saf. 2020;15(1):4–12. doi: 10.2174/1574886314666191004092520
   PubMed Web of Science ®Google Scholar
• Kocadal K, Saygi S, Alkas FB, et al. Drug-associated cardiovascular risks: a retrospective evaluation of withdrawn drugs. North Clin Istanb. 2019;6(2):196–202. doi: 10.14744/nci.2018.44977
   PubMed Web of Science ®Google Scholar
• Babcock JJ, Li M. hERG channel function: beyond long QT. Acta Pharmacol Sin. 2013 Mar;34(3):329–335. doi: 10.1038/aps.2013.6
   PubMed Web of Science ®Google Scholar
• Rampe D, Brown AM. A history of the role of the hERG channel in cardiac risk assessment. J Pharmacol Toxicol Methods. 2013 Jul;68(1):13–22. doi: 10.1016/j.vascn.2013.03.005
   PubMed Web of Science ®Google Scholar
• John VH, Dale TJ, Hollands EC, et al. Novel 384-well population patch clamp electrophysiology assays for Ca2±activated K+ channels. J Biomol Screen. 2007 Feb;12(1):50–60. doi: 10.1177/1087057106294920
   PubMedGoogle Scholar
• Nomura T. In vitro patch-clamp. Methods Mol Biol. 2024;2794:221–244.
   PubMedGoogle Scholar
• Miranda P, Manso DG, Barros F, et al. FRET with multiply labeled HERG K(+) channels as a reporter of the in vivo coarse architecture of the cytoplasmic domains. Bba-Mol Cell Res. 2008 Oct;1783(10):1681–1699. doi: 10.1016/j.bbamcr.2008.06.009
   PubMed Web of Science ®Google Scholar
• Margulis M, Sorota S, Chu I, et al. Protein binding-dependent decreases in hERG channel blocker potency assessed by whole-cell voltage clamp in serum. J Cardiovasc Pharmacol. 2010 Apr;55(4):368–376. doi: 10.1097/FJC.0b013e3181d2ce39
   PubMed Web of Science ®Google Scholar
• Weaver CD. Thallium flux assay for measuring the activity of monovalent cation channels and transporters. Methods Mol Biol. 2018;1684:105–114.
   PubMedGoogle Scholar
• Schmalhofer WA, Swensen AM, Thomas BS, et al. A pharmacologically validated, high-capacity, functional thallium flux assay for the human ether-a-go-go related gene potassium channel. Assay Drug Dev Technol. 2010 Dec;8(6):714–726. doi: 10.1089/adt.2010.0351
   PubMed Web of Science ®Google Scholar
• Zhao J, Xia M. Cell-based hERG channel inhibition assay in high-throughput format. Methods Mol Biol. 2022;2474:21–28.
   PubMedGoogle Scholar
• Rogers M, Obergrussberger A, Kondratskyi A, et al. Using automated patch clamp electrophysiology platforms in ion channel drug discovery: an industry perspective. Expert Opin Drug Discov. 2024 May;19(5):523–535. doi: 10.1080/17460441.2024.2329104
   PubMed Web of Science ®Google Scholar
• Vicente J. Update on the ECG component of the CiPA initiative. J Electrocardiol. 2018 Nov;51(6S):S98–S102. doi: 10.1016/j.jelectrocard.2018.08.003
   PubMedGoogle Scholar
• Priest BT, Bell IM, Garcia ML. Role of hERG potassium channel assays in drug development. Channels (Austin). 2008 Mar;2(2):87–93. doi: 10.4161/chan.2.2.6004
   PubMed Web of Science ®Google Scholar
• Kim T, Chung KC, Park H. Derivation of highly predictive 3D-QSAR models for hERG channel blockers based on the quantum artificial neural network algorithm. Pharmaceuticals (Basel). 2023 Oct 24;16(11):1509. doi: 10.3390/ph16111509
   PubMedGoogle Scholar
• Karim A, Lee M, Balle T, et al. CardioTox net: a robust predictor for hERG channel blockade based on deep learning meta-feature ensembles. J Cheminform. 2021 Aug 16;13(1):60. doi: 10.1186/s13321-021-00541-z
   PubMed Web of Science ®Google Scholar
• Ryu JY, Lee MY, Lee JH, et al. DeepHIT: a deep learning framework for prediction of hERG-induced cardiotoxicity. Bioinformatics. 2020 May 1;36(10):3049–3055. doi: 10.1093/bioinformatics/btaa075
   PubMed Web of Science ®Google Scholar
• Delre P, Lavado GJ, Lamanna G, et al. Ligand-based prediction of hERG-mediated cardiotoxicity based on the integration of different machine learning techniques. Front Pharmacol. 2022;13:951083. doi: 10.3389/fphar.2022.951083
   PubMed Web of Science ®Google Scholar
• Krishna S, Borrel A, Huang R, et al. High-throughput chemical screening and structure-based models to predict hERG inhibition. Biology (Basel). 2022 Jan 28;11(2):209. doi: 10.3390/biology11020209
   PubMedGoogle Scholar
• Lee HM, Yu MS, Kazmi SR, et al. Computational determination of hERG-related cardiotoxicity of drug candidates. BMC Bioinformatics. 2019 May 29;20(Suppl 10):250. doi: 10.1186/s12859-019-2814-5
   PubMedGoogle Scholar
• Braga RC, Alves VM, Silva MF, et al. Pred-hERG: a novel web-accessible computational tool for predicting cardiac toxicity. Mol Inform. 2015 Oct;34(10):698–701. doi: 10.1002/minf.201500040
   PubMed Web of Science ®Google Scholar
• Li X, Zhang Y, Li H, et al. Modeling of the hERG K+ channel blockage using online chemical database and modeling environment (OCHEM). Mol Inform. 2017 Dec;36(12). doi: 10.1002/minf.201700074
   Web of Science ®Google Scholar
• Siramshetty VB, Nguyen DT, Martinez NJ, et al. Critical assessment of artificial intelligence methods for prediction of hERG channel inhibition in the “Big Data” Era. J Chem Inf Model. 2020 Dec 28;60(12):6007–6019. doi: 10.1021/acs.jcim.0c00884
   PubMed Web of Science ®Google Scholar
• Kim H, Nam H. hERG-att: self-attention-based deep neural network for predicting hERG blockers. Comput Biol Chem. 2020 May 19;87:107286. doi: 10.1016/j.compbiolchem.2020.107286
   PubMed Web of Science ®Google Scholar
• Kim H, Park M, Lee I, et al. BayeshERG: a robust, reliable and interpretable deep learning model for predicting hERG channel blockers. Brief Bioinform. 2022 Jul 18;23(4). doi: 10.1093/bib/bbac211
   Web of Science ®Google Scholar
• Cai C, Guo P, Zhou Y, et al. Deep learning-based prediction of drug-induced cardiotoxicity. J Chem Inf Model. 2019 Mar 25;59(3):1073–1084. doi: 10.1021/acs.jcim.8b00769
   PubMed Web of Science ®Google Scholar
• Creanza TM, Delre P, Ancona N, et al. Structure-based prediction of hERG-related cardiotoxicity: a benchmark study. J Chem Inf Model. 2021 Sep 27;61(9):4758–4770. doi: 10.1021/acs.jcim.1c00744
   PubMed Web of Science ®Google Scholar
• Liu M, Zhang L, Li S, et al. Prediction of hERG potassium channel blockage using ensemble learning methods and molecular fingerprints. Toxicol Lett. 2020 Oct 10;332:88–96. doi: 10.1016/j.toxlet.2020.07.003
   PubMed Web of Science ®Google Scholar
• Zhang C, Zhou Y, Gu S, et al. In silico prediction of hERG potassium channel blockage by chemical category approaches. Toxicol Res (Camb). 2016 Mar 1;5(2):570–582. doi: 10.1039/C5TX00294J
   PubMed Web of Science ®Google Scholar
• Wang S, Sun H, Liu H, et al. ADMET evaluation in drug discovery. 16. Predicting hERG blockers by combining multiple pharmacophores and machine learning approaches. Mol Pharm. 2016 Aug 1;13(8):2855–2866. doi: 10.1021/acs.molpharmaceut.6b00471
   PubMed Web of Science ®Google Scholar
• Sun H, Huang R, Xia M, et al. Prediction of hERG liability – using SVM classification, bootstrapping and jackknifing. Mol Inform. 2017 Apr;36(4). doi: 10.1002/minf.201600126
   Web of Science ®Google Scholar
• Chavan S, Abdelaziz A, Wiklander JG, et al. A k-nearest neighbor classification of hERG K(+) channel blockers. J Comput Aided Mol Des. 2016 Mar;30(3):229–236. doi: 10.1007/s10822-016-9898-z
   PubMed Web of Science ®Google Scholar
• Ogura K, Sato T, Yuki H, et al. Support vector machine model for hERG inhibitory activities based on the integrated hERG database using descriptor selection by NSGA-II. Sci Rep. 2019 Aug 21;9(1):12220. doi: 10.1038/s41598-019-47536-3
   PubMedGoogle Scholar
• Racz A, Bajusz D, Miranda-Quintana RA, et al. Machine learning models for classification tasks related to drug safety. Mol Divers. 2021 Aug;25(3):1409–1424. doi: 10.1007/s11030-021-10239-x
   PubMed Web of Science ®Google Scholar
• Lanevskij K, Didziapetris R, Sazonovas A. Physicochemical QSAR analysis of hERG inhibition revisited: towards a quantitative potency prediction. J Comput Aided Mol Des. 2022 Dec;36(12):837–849. doi: 10.1007/s10822-022-00483-0
   PubMed Web of Science ®Google Scholar
• Singh AV, Varma M, Rai M, et al. Advancing predictive risk assessment of chemicals via integrating machine learning, computational modeling, and chemical/nano-quantitative structure-activity relationship approaches. Adv Intell Syst-Ger. 2024 Apr;6(4). doi: 10.1002/aisy.202300366
   Google Scholar
• Karim A, Lee M, Balle T, et al. CardioTox training data from: CardioTox net: a robust predictor for hERG channel blockade based on deep learning meta-feature ensembles. 2021 Aug 16 [cited 2023 Jun 21]. Available from: https://github.com/Abdulk084/CardioTox/blob/master/data/train_validation_cardio_tox_data.tar.xz
   Google Scholar
• Online SMILES Translator and Structure File Generator. [cited 2023 Jun 28]. Available from: https://cactus.nci.nih.gov/translate/
   Google Scholar
• Karim A, Lee M, Balle T, et al. CardioTox external validation dataset-1 from: CardioTox net: a robust predictor for hERG channel blockade based on deep learning meta-feature ensembles. 2021 Aug 16 [cited 2023 Jun 21]. Available from: https://github.com/Abdulk084/CardioTox/blob/master/data/external_test_set_pos.csv
   Google Scholar
• Karim A, Lee M, Balle T, et al. CardioTox external validation dataset-2 from: CardioTox net: a robust predictor for hERG channel blockade based on deep learning meta-feature ensembles. 2021 Aug 16 [cited 2023 Jun 21]. Available from: https://github.com/Abdulk084/CardioTox/blob/master/data/external_test_set_new.csv
   Google Scholar
• Mold2. [cited 2023 Jun 28]. Available from: https://www.fda.gov/science-research/bioinformatics-tools/mold2
   Google Scholar
• Hong H, Xie Q, Ge W, et al. Mold(2), molecular descriptors from 2D structures for chemoinformatics and toxicoinformatics. J Chem Inf Model. 2008 Jul;48(7):1337–1344. doi: 10.1021/ci800038f
   PubMed Web of Science ®Google Scholar
• Hong H, Liu J, Ge W, et al. Mold2 descriptors facilitate development of machine learning and deep learning models for predicting toxicity of chemicals. In: Hong H, editor. Machine learning and deep learning in computational toxicology; Computational methods in engineering & the sciences; 2023. Cham: Springer International Publishing; p. 297–321. doi:10.1007/978-3-031-20730-3_12
   Google Scholar
• Chen M, Hong H, Fang H, et al. Quantitative structure-activity relationship models for predicting drug-induced liver injury based on FDA-approved drug labeling annotation and using a large collection of drugs. Toxicol Sci. 2013 Nov;136(1):242–249. doi: 10.1093/toxsci/kft189
   PubMed Web of Science ®Google Scholar
• Shen J, Xu L, Fang H, et al. EADB: an estrogenic activity database for assessing potential endocrine activity. Toxicol Sci. 2013 Oct;135(2):277–291. doi: 10.1093/toxsci/kft164
   PubMed Web of Science ®Google Scholar
• Ng HW, Doughty SW, Luo H, et al. Development and validation of decision forest model for estrogen receptor binding prediction of chemicals using large data sets. Chem Res Toxicol. 2015 Dec 21;28(12):2343–2351. doi: 10.1021/acs.chemrestox.5b00358
   PubMed Web of Science ®Google Scholar
• Sakkiah S, Selvaraj C, Gong P, et al. Development of estrogen receptor beta binding prediction model using large sets of chemicals. Oncotarget. 2017 Nov 3;8(54):92989–93000. doi: 10.18632/oncotarget.21723
   PubMedGoogle Scholar
• Hong H, Rua D, Sakkiah S, et al. Consensus modeling for prediction of estrogenic activity of ingredients commonly used in sunscreen products. Int J Environ Res Public Health. 2016 Sep 29;13(10):958. doi: 10.3390/ijerph13100958
   PubMed Web of Science ®Google Scholar
• Hong H, Harvey BG, Palmese GR, et al. Experimental data extraction and in silico prediction of the estrogenic activity of renewable replacements for bisphenol a. Int J Environ Res Public Health. 2016 Jul 12;13(7):705. doi: 10.3390/ijerph13070705
   PubMed Web of Science ®Google Scholar
• Hong H, Shen J, Ng HW, et al. A rat alpha-fetoprotein binding activity prediction model to facilitate assessment of the endocrine disruption potential of environmental chemicals. Int J Environ Res Public Health. 2016 Mar 25;13(4):372. doi: 10.3390/ijerph13040372
   PubMed Web of Science ®Google Scholar
• Hong H, Thakkar S, Chen M, et al. Development of decision forest models for prediction of drug-induced liver injury in humans using a large set of FDA-approved drugs. Sci Rep. 2017 Dec 11;7(1):17311. doi: 10.1038/s41598-017-17701-7
   PubMedGoogle Scholar
• Liu J, Guo W, Dong F, et al. Machine learning models for rat multigeneration reproductive toxicity prediction. Front Pharmacol. 2022;13:1018226. doi: 10.3389/fphar.2022.1018226
   PubMed Web of Science ®Google Scholar
• Liu J, Xu L, Guo W, et al. Developing a SARS-CoV-2 main protease binding prediction random forest model for drug repurposing for COVID-19 treatment. Exp Biol Med (Maywood). 2023 Nov;248(21):1927–1936. doi: 10.1177/15353702231209413
   PubMed Web of Science ®Google Scholar
• Godden JW, Stahura FL, Bajorath J. Variability of molecular descriptors in compound databases revealed by Shannon entropy calculations. J Chem Inf Comput Sci. 2000 May;40(3):796–800. doi: 10.1021/ci000321u
   PubMedGoogle Scholar
• Shi L, Tong W, Fang H, et al. An integrated “4-phase” approach for setting endocrine disruption screening priorities–phase I and II predictions of estrogen receptor binding affinity. SAR QSAR Environ Res. 2002 Mar;13(1):69–88. doi: 10.1080/10629360290002235
   PubMed Web of Science ®Google Scholar
• Hong H, Tong W, Xie Q, et al. An in silico ensemble method for lead discovery: decision forest. SAR QSAR Environ Res. 2005 Aug;16(4):339–347. doi: 10.1080/10659360500203022
   PubMed Web of Science ®Google Scholar
• Luo H, Ye H, Ng HW, et al. Machine learning methods for predicting HLA-peptide binding activity. Bioinform Biol Insights. 2015;9(Suppl 3):21–29. doi: 10.4137/BBI.S29466
   PubMedGoogle Scholar
• Ye H, Luo H, Ng HW, et al. Applying network analysis and Nebula (neighbor-edges based and unbiased leverage algorithm) to ToxCast data. Environ Int. 2016 Apr;89-90:81–92. doi: 10.1016/j.envint.2016.01.010
   PubMed Web of Science ®Google Scholar
• Idakwo G, Thangapandian S, Luttrell J, et al. Structure-activity relationship-based chemical classification of highly imbalanced Tox21 datasets. J Cheminform. 2020 Oct 27;12(1):66. doi: 10.1186/s13321-020-00468-x
   PubMed Web of Science ®Google Scholar
• Huang Y, Li X, Xu S, et al. Quantitative structure-activity relationship models for predicting inflammatory potential of metal oxide nanoparticles. Environ Health Perspect. 2020 Jun;128(6):67010. doi: 10.1289/EHP6508
   PubMed Web of Science ®Google Scholar
• Wang Z, Chen J, Hong H. Developing QSAR models with defined applicability domains on PPARgamma binding affinity using large data sets and machine learning algorithms. Environ Sci Technol. 2021 May 18;55(10):6857–6866. doi: 10.1021/acs.est.0c07040
   PubMed Web of Science ®Google Scholar
• Guo W, Liu J, Dong F, et al. Deep learning models for predicting gas adsorption capacity of nanomaterials. Nanomaterials (Basel). 2022 Sep 27;12(19):3376. doi: 10.3390/nano12193376
   PubMedGoogle Scholar
• Liu J, Guo W, Sakkiah S, et al. Machine learning models for predicting liver toxicity. Methods Mol Biol. 2022;2425:393–415.
   PubMedGoogle Scholar
• Ji Z, Guo W, Wood EL, et al. Machine learning models for predicting cytotoxicity of nanomaterials. Chem Res Toxicol. 2022 Feb 21;35(2):125–139. doi: 10.1021/acs.chemrestox.1c00310
   PubMed Web of Science ®Google Scholar
• Tang W, Liu W, Wang Z, et al. Machine learning models on chemical inhibitors of mitochondrial electron transport chain. J Hazard Mater. 2022 Mar 15;426:128067. doi: 10.1016/j.jhazmat.2021.128067
   PubMed Web of Science ®Google Scholar
• Guo W, Liu J, Dong F, et al. Review of machine learning and deep learning models for toxicity prediction. Exp Biol Med (Maywood). 2023 Nov;248(21):1952–1973. doi: 10.1177/15353702231209421
   PubMed Web of Science ®Google Scholar
• Python. Available from: https://www.python.org/downloads/release/python-385/
   Google Scholar
• Scikit-learn. Available from: https://scikit-learn.org/stable/whats_new/v1.2.html
   Google Scholar
• PyTorch. Available from: https://pytorch.org/get-started/pytorch-2.0/
   Google Scholar
• Cover TM, Hart PE. Nearest neighbor pattern classification. IEEE T Inf Theory. 1967;13(1):21±. doi: 10.1109/TIT.1967.1053964
   Web of Science ®Google Scholar
• Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995 Sep;20(3):273–297. doi: 10.1007/BF00994018
   Web of Science ®Google Scholar
• Breiman L. Random forests. Mach Learn. 2001 Oct;45(1):5–32. doi: 10.1023/A:1010933404324
   Web of Science ®Google Scholar
• Rosenblatt F. The perceptron – a probabilistic model for information-storage and organization in the brain. Psychol Rev. 1958;65(6):386–408. doi: 10.1037/h0042519
   PubMed Web of Science ®Google Scholar
• Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput. 1997 Nov 15;9(8):1735–1780. doi: 10.1162/neco.1997.9.8.1735
   PubMed Web of Science ®Google Scholar
• Vincent P, Larochelle H, Lajoie I, et al. Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J Mach Learn Res. 2010 Dec;11:3371–3408.
   Web of Science ®Google Scholar
• Opitz D, Maclin R. Popular ensemble methods: An empirical study. J Artif Intel Res. 1999;11(1):169–198. doi: 10.1613/jair.614
   Google Scholar
• Chandrasekar V, Ansari MY, Singh AV, et al. Investigating the use of machine learning models to understand the drugs permeability across placenta. IEEE Access. 2023;11:52726–52739. doi: 10.1109/ACCESS.2023.3272987
   Web of Science ®Google Scholar
• Kar S, Roy K, Leszczynski J. Applicability domain: a step toward confident predictions and decidability for QSAR modeling. Methods Mol Biol. 2018;1800:141–169.
   PubMedGoogle Scholar
• Singh AV, Shelar A, Rai M, et al. Harmonization risks and rewards: nano-QSAR for agricultural nanomaterials. J Agric Food Chem. 2024 Feb 14;72(6):2835–2852. doi: 10.1021/acs.jafc.3c06466
   PubMed Web of Science ®Google Scholar
• Wisniowska B, Mendyk A, Szlek J, et al. Enhanced QSAR models for drug-triggered inhibition of the main cardiac ion currents. J Appl Toxicol. 2015 Sep;35(9):1030–1039. doi: 10.1002/jat.3095
   PubMed Web of Science ®Google Scholar
• Singh AV, Varma M, Laux P, et al. Artificial intelligence and machine learning disciplines with the potential to improve the nanotoxicology and nanomedicine fields: a comprehensive review. Arch Toxicol. 2023 Apr;97(4):963–979. doi: 10.1007/s00204-023-03471-x
   PubMed Web of Science ®Google Scholar
• Park JS, Jeon JY, Yang JH, et al. Introduction to in silico model for proarrhythmic risk assessment under the CiPA initiative. Transl Clin Pharmacol. 2019 Mar;27(1):12–18. doi: 10.12793/tcp.2019.27.1.12
   PubMedGoogle Scholar
• Hong HX, Xu L, Liu J, et al. Technical reproducibility of genotyping SNP arrays used in genome-wide association studies. PLOS ONE. 2012 Sep 7;7(9):e44483. doi: 10.1371/journal.pone.0044483
   PubMed Web of Science ®Google Scholar
• Moriwaki H, Tian YS, Kawashita N, et al. Mordred: a molecular descriptor calculator. J Cheminf. 2018 Feb 6;10(1). doi: 10.1186/s13321-018-0258-y
   PubMedGoogle Scholar
• Li Z, Huang RL, Xia MH, et al. Fingerprinting interactions between proteins and ligands for facilitating machine learning in drug discovery. Biomol. 2024 Jan;14(1):72. doi: 10.3390/biom14010072
   Google Scholar