{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T13:06:49Z","timestamp":1788959209115,"version":"build-2803163510"},"reference-count":41,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T00:00:00Z","timestamp":1788912000000},"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. Pharmacol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Bile salt export pump (BSEP) inhibition is an important mechanistic risk factor for cholestatic drug-induced liver injury. Quantitative structure-activity relationship (QSAR) models may support early safety screening, but practical BSEP datasets are often limited in size and imbalanced.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We evaluated a machine-learning workflow combining molecular descriptors and fingerprints, Boruta feature selection, and class-imbalance training. Multiple algorithms were assessed using cross-validation, a held-out test set, and an independent external validation set composed of compounds synthesized in an industrial research setting.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Tree-based models generally showed the strongest external performance. In particular, models using RDKit2D descriptors achieved Matthews correlation coefficients (MCC) above 0.65 and a ROC-AUC of 0.87, while several other tree-based model representation combinations achieved MCC values around 0.45-0.50. Boruta markedly reduced feature dimensionality, but its effect on predictive performance depended on both the molecular representation and learning algorithm; it was most beneficial for selected high-dimensional fingerprints and tree-based models, while offering limited or unfavorable effects in some support vector machine settings. Applicability-domain and SHAP analyses further revealed representation-dependent generalization patterns and highlighted contributions from lipophilicity, polarity, solubility, and molecular complexity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Overall, the workflow maintained useful predictive performance under small and imbalanced conditions and provides a practical framework for early-stage BSEP risk screening.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fphar.2026.1911283","type":"journal-article","created":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T12:16:35Z","timestamp":1788956195000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Machine learning prediction of bile salt export pump inhibition with Boruta feature selection in imbalanced data"],"prefix":"10.3389","volume":"17","author":[{"given":"Bailang","family":"Liu","sequence":"first","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yong","family":"Liu","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hui","family":"Yang","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Li","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shanshan","family":"Yin","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Junnan","family":"Wang","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhikui","family":"Cheng","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Zhao","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuejun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shu","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Infectious Diseases, Wuhan University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lifei","family":"Liu","sequence":"additional","affiliation":[{"name":"Humanwell Healthcare (Group) Co, Ltd.","place":["Wuhan, Hubei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"21853","DOI":"10.1021\/acsomega.3c01583","article-title":"Using a graph convolutional neural network model to identify bile salt export pump inhibitors","volume":"8","author":"AbdulHameed","year":"2023","journal-title":"ACS Omega"},{"key":"B2","doi-asserted-by":"publisher","first-page":"101944","DOI":"10.1016\/j.softx.2024.101944","article-title":"Scikit-fingerprints: easy and efficient computation of molecular fingerprints in Python","volume":"28","author":"Adamczyk","year":"2024","journal-title":"SoftwareX"},{"key":"B3","doi-asserted-by":"publisher","first-page":"715","DOI":"10.1016\/j.therap.2018.02.004","article-title":"Safety data and withdrawal of hepatotoxic drugs","volume":"76","author":"Babai","year":"2021","journal-title":"Therapie"},{"key":"B4","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Machine Learning"},{"key":"B5","doi-asserted-by":"publisher","first-page":"71","DOI":"10.3727\/105221618X15156018385515","article-title":"Bile acid metabolism in liver pathobiology","volume":"18","author":"Chiang","year":"2018","journal-title":"Gene Expr."},{"key":"B6","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1007\/BF00994018","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn"},{"key":"B7","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/TIT.1967.1053964","article-title":"Nearest neighbor pattern classification","volume":"13","author":"Cover","year":"1967","journal-title":"IEEE Trans. Inf. Theory"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1548","DOI":"10.3390\/molecules30071548","article-title":"Predicting the toxicity of drug molecules with selecting effective descriptors using a Binary Ant Colony Optimization (BACO) feature selection approach","volume":"30","author":"Dan","year":"2025","journal-title":"Molecules"},{"key":"B9","doi-asserted-by":"publisher","first-page":"e03069","DOI":"10.1016\/j.sciaf.2025.e03069","article-title":"2D and 3D QSAR-based machine learning models for predicting pyrazole corrosion inhibitors for mild steel in HCl","volume":"30","author":"El-Idrissi","year":"2025","journal-title":"Sci. Afr."},{"key":"B10","doi-asserted-by":"publisher","first-page":"1924","DOI":"10.1021\/ci050413p","article-title":"QSAR - how good is it in practice? Comparison of descriptor sets on an unbiased cross section of corporate data sets","volume":"46","author":"Gedeck","year":"2006","journal-title":"J. Chem. Inf. Model."},{"key":"B11","doi-asserted-by":"publisher","first-page":"252","DOI":"10.1021\/mp060004w","article-title":"High-speed screening and QSAR analysis of human ATP-binding cassette transporter ABCB11 (bile salt export pump) to predict drug-induced intrahepatic cholestasis","volume":"3","author":"Hirano","year":"2006","journal-title":"Mol. Pharm."},{"key":"B12","doi-asserted-by":"publisher","first-page":"17311","DOI":"10.1038\/s41598-017-17701-7","article-title":"Development of decision forest models for prediction of drug-induced liver injury in humans using A large set of FDA-approved drugs","volume":"7","author":"Hong","year":"2017","journal-title":"Sci. Rep."},{"key":"B13","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1007\/978-1-4939-7677-5_5","article-title":"Quantitative structure\u2013activity relationship models for predicting risk of drug-induced liver injury in humans","volume-title":"Drug-Induced Liver Toxicity","author":"Hong","year":"2018"},{"key":"B14","doi-asserted-by":"publisher","first-page":"104234","DOI":"10.1016\/j.phrs.2019.04.018","article-title":"Drugs and hepatic transporters: a review","volume":"154","author":"Jetter","year":"2020","journal-title":"Pharmacol. Res."},{"key":"B15","doi-asserted-by":"publisher","first-page":"865","DOI":"10.1007\/s11427-023-2353-0","article-title":"Bile acid signaling in the regulation of whole body metabolic and immunological homeostasis","volume":"67","author":"Jia","year":"2024","journal-title":"Sci. China Life Sci."},{"key":"B16","doi-asserted-by":"publisher","first-page":"916","DOI":"10.1002\/cpt.1222","article-title":"Can bile salt export pump inhibition testing in drug discovery and development reduce liver injury risk? An international transporter consortium perspective","volume":"104","author":"Kenna","year":"2018","journal-title":"Clin. Pharmacol. Ther."},{"key":"B17","doi-asserted-by":"publisher","first-page":"D1202","DOI":"10.1093\/nar\/gkv951","article-title":"PubChem substance and compound databases","volume":"44","author":"Kim","year":"2016","journal-title":"Nucleic Acids Res."},{"key":"B18","doi-asserted-by":"publisher","first-page":"1294","DOI":"10.1016\/j.tiv.2011.11.002","article-title":"BSEP inhibition: in vitro screens to assess cholestatic potential of drugs","volume":"26","author":"Kis","year":"2012","journal-title":"Toxicol Vitro"},{"key":"B19","doi-asserted-by":"publisher","first-page":"271","DOI":"10.3233\/FI-2010-288","article-title":"Boruta \u2013 a system for feature selection","volume":"101","author":"Kursa","year":"2010","journal-title":"Fundam. Inf."},{"key":"B20","volume-title":"rdkit\/rdkit: 2025_03_3 (Q1 2025) Release","author":"Landrum","year":"2025"},{"key":"B21","doi-asserted-by":"publisher","first-page":"10359","DOI":"10.3389\/ebm.2025.10359","article-title":"Developing predictive models for \u00b5 opioid receptor binding using machine learning and deep learning techniques","volume":"250","author":"Liu","year":"2025","journal-title":"Exp. Biol. Med. (Maywood)"},{"key":"B22","article-title":"A unified approach to interpreting model predictions","volume-title":"Advances in Neural Information Processing Systems","author":"Lundberg","year":"2017"},{"key":"B23","doi-asserted-by":"publisher","first-page":"587","DOI":"10.1021\/acs.jcim.0c00950","article-title":"Machine learning models to predict inhibition of the bile salt export pump","volume":"61","author":"McLoughlin","year":"2021","journal-title":"J. Chem. Inf. Model"},{"key":"B24","doi-asserted-by":"publisher","first-page":"163","DOI":"10.1021\/acs.molpharmaceut.5b00594","article-title":"Flagging drugs that inhibit the bile salt export pump","volume":"13","author":"Montanari","year":"2016","journal-title":"Mol. Pharm."},{"key":"B25","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1093\/toxsci\/kft176","article-title":"A multifactorial approach to hepatobiliary transporter assessment enables improved therapeutic compound development","volume":"136","author":"Morgan","year":"2013","journal-title":"Toxicol. Sci."},{"key":"B26","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1186\/s13321-018-0258-y","article-title":"Mordred: a molecular descriptor calculator","volume":"10","author":"Moriwaki","year":"2018","journal-title":"J. Cheminform"},{"key":"B27","doi-asserted-by":"publisher","first-page":"1340860","DOI":"10.3389\/ftox.2023.1340860","article-title":"Computational models for predicting liver toxicity in the deep learning era","volume":"5","author":"Mostafa","year":"2023","journal-title":"Front. Toxicol."},{"key":"B28","doi-asserted-by":"publisher","first-page":"72","DOI":"10.14744\/hf.2024.2024.0003","article-title":"Drug-induced liver injury: diagnosis, management and the role of liver transplantation","volume":"6","author":"Ozturk","year":"2024","journal-title":"Hepatol. Forum"},{"key":"B29","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"B30","doi-asserted-by":"publisher","first-page":"202","DOI":"10.37349\/edd.2023.00027","article-title":"Drug-induced cholestasis: causative agents and challenges in diagnosis and management","volume":"2","author":"Pinazo-Bandera","year":"2023","journal-title":"Explor Dig. Dis."},{"key":"B31","volume-title":"CatBoost: Unbiased Boosting with Categorical Features","author":"Prokhorenkova","year":"2019"},{"key":"B32","doi-asserted-by":"publisher","first-page":"101641","DOI":"10.1016\/j.clinre.2021.101641","article-title":"Regulatory mechanisms of the bile salt export pump (BSEP\/ABCB11) and its role in related diseases","volume":"45","author":"Ren","year":"2021","journal-title":"Clin. Res. Hepatol. Gastroenterol."},{"key":"B33","doi-asserted-by":"publisher","first-page":"100027","DOI":"10.1016\/j.ailsci.2021.100027","article-title":"Identification of bile salt export pump inhibitors using machine learning: predictive safety from an industry perspective","volume":"1","author":"Rodr\u00edguez-P\u00e9rez","year":"2021","journal-title":"Artif. Intell. Life Sci."},{"key":"B34","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1186\/s12859-023-05176-5","article-title":"PredAOT: a computational framework for prediction of acute oral toxicity based on multiple random forest models","volume":"24","author":"Ryu","year":"2023","journal-title":"BMC Bioinforma."},{"key":"B35","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1186\/s13287-025-04139-2","article-title":"Advances in liver organoids: replicating hepatic complexity for toxicity assessment and disease modeling","volume":"16","author":"Shao","year":"2025","journal-title":"Stem Cell Res. & Ther."},{"key":"B36","doi-asserted-by":"publisher","first-page":"4908","DOI":"10.1038\/s41598-023-31169-8","article-title":"Accurate clinical toxicity prediction using multi-task deep neural nets and contrastive molecular explanations","volume":"13","author":"Sharma","year":"2023","journal-title":"Sci. Rep."},{"key":"B37","doi-asserted-by":"publisher","first-page":"4323","DOI":"10.1038\/s41467-023-40064-9","article-title":"A preclinical secondary pharmacology resource illuminates target-adverse drug reaction associations of marketed drugs","volume":"14","author":"Sutherland","year":"2023","journal-title":"Nat. Commun."},{"key":"B38","doi-asserted-by":"publisher","first-page":"2332","DOI":"10.1124\/dmd.112.047068","article-title":"Mitigating the inhibition of human bile salt export pump by drugs: opportunities provided by physicochemical property modulation, in silico modeling, and structural modification","volume":"40","author":"Warner","year":"2012","journal-title":"Drug Metabolism Dispos."},{"key":"B39","doi-asserted-by":"publisher","first-page":"3112","DOI":"10.3390\/molecules27103112","article-title":"An algorithm framework for drug-induced liver injury prediction based on genetic algorithm and ensemble learning","volume":"27","author":"Yan","year":"2022","journal-title":"Molecules"},{"key":"B40","doi-asserted-by":"publisher","first-page":"3370","DOI":"10.1021\/acs.jcim.9b00237","article-title":"Analyzing learned molecular representations for property prediction","volume":"59","author":"Yang","year":"2019","journal-title":"J. Chem. Inf. Model."},{"key":"B41","doi-asserted-by":"publisher","first-page":"998","DOI":"10.1016\/j.jhep.2023.01.019","article-title":"A human liver organoid screening platform for DILI risk prediction","volume":"78","author":"Zhang","year":"2023","journal-title":"J. Hepatol."}],"container-title":["Frontiers in Pharmacology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphar.2026.1911283\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T12:16:38Z","timestamp":1788956198000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphar.2026.1911283\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,9]]},"references-count":41,"alternative-id":["10.3389\/fphar.2026.1911283"],"URL":"https:\/\/doi.org\/10.3389\/fphar.2026.1911283","relation":{},"ISSN":["1663-9812"],"issn-type":[{"value":"1663-9812","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,9]]},"article-number":"1911283"}}