{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,8]],"date-time":"2026-09-08T14:06:43Z","timestamp":1788876403046,"version":"build-2803163510"},"reference-count":36,"publisher":"American Chemical Society (ACS)","issue":"35","license":[{"start":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T00:00:00Z","timestamp":1787702400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,8]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Early stage preclinical drug formulation development can be impacted by limited materials and time, especially when making rapid, data-driven decisions is critical. An important parameter to consider is dose linearity, as this can guide the selection between determining if a conventional or enabled formulation strategy is necessary to achieve the expected target exposure as the dose is escalated. This work explores the application of machine learning models to predict dose linearity for compounds administered to the mouse species based on the molecular properties. We utilize a diverse set of models, including Random Forest, XGBoost, LASSO, Neural Networks, and TabPFN, using two distinct data sets for training: a full data set including measured and calculated descriptors and a more accessible data set containing only the RDKit descriptors. Cross-validation was applied with random and scaffold splits, and model performance was evaluated based on a test data set containing 10% of the full data set that was held out during training. Our analysis indicated that the ML models trained with the full descriptors were comparable to the SMILES-only derived descriptors with RDKit. Additionally, while most models achieved comparable performance, LASSO was prioritized due to the interpretability, feature selection, and resistance to overfitting smaller data sets. With a random split and RDKit-only descriptors, the LASSO model resulted in 90.9% precision in the test set and 88.0% accuracy. Precision was optimized to minimize false positives, as incorrectly determining that a compound would not be less-than-dose proportional, where a conventional vehicle would be sufficient, has a higher cost than false negatives. To further evaluate the model\u2019s capacity to apply to novel chemical compounds, a scaffold split was tested. A gap in performance was seen with a precision of 89.3% in training and 80.9% in testing, highlighting the challenges in predicting outcomes with new, chemically unique architectures. A significant finding of this work is the robust performance of the models trained solely on accessible RDKit descriptors, demonstrating that a tool can be developed without more resource-intensive measured data. Overall, this research focuses on a data-driven approach based on dose linearity prediction for whether a conventional, nonsolubilizing vehicle is sufficient or if an enabled formulation would be required to achieve the target exposures. The application to forecast a compound\u2019s dose linearity risk with only the chemical structures provides a valuable tool to guide early formulation strategy and accelerate compound evaluation and preclinical development during the drug discovery phase.<\/jats:p>","DOI":"10.1021\/acsomega.6c05839","type":"journal-article","created":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T07:52:40Z","timestamp":1788249160000},"page":"53186-53195","source":"Crossref","is-referenced-by-count":0,"title":["Machine Learning-Driven\nPrediction of Dose-Linear\nPharmacokinetics: Utilizing Molecular Descriptors to Guide Formulation\nStrategy"],"prefix":"10.1021","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6802-885X","authenticated-orcid":true,"given":"Elizabeth S.","family":"Levy","sequence":"first","affiliation":[{"name":"Genentech Inc , , , ,","place":["South San Francisco, California, United States, 94080"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Karen E.","family":"Samy","sequence":"additional","affiliation":[{"name":"Genentech Inc , , , ,","place":["South San Francisco, California, United States, 94080"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrea","family":"Anelli","sequence":"additional","affiliation":[{"name":"Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd. , , ,","place":["Basel, Switzerland, 4070"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dennis H.","family":"Leung","sequence":"additional","affiliation":[{"name":"Genentech Inc , , , ,","place":["South San Francisco, California, United States, 94080"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,26]]},"reference":[{"key":"2026090809073964100_cit1","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1016\/j.bcp.2013.09.007","article-title":"Pharmacokinetics","volume":"87","author":"Fan","year":"2014","journal-title":"Biochem. Pharmacol."},{"key":"2026090809073964100_cit2","doi-asserted-by":"publisher","first-page":"126","DOI":"10.12793\/tcp.2020.28.e14","article-title":"Predicting\nHuman Pharmacokinetics\nfrom Preclinical Data: Absorption","volume":"28","author":"Yim","year":"2020","journal-title":"Transl. Clin.\nPharmacol."},{"key":"2026090809073964100_cit3","doi-asserted-by":"publisher","DOI":"10.1155\/2013\/848043","article-title":"Formulation Strategies\nto Improve\nthe Bioavailability of Poorly Absorbed Drugs with Special Emphasis\non Self-Emulsifying Systems","volume":"2013","author":"Gupta","year":"2013","journal-title":"ISRN Pharmaceutics"},{"key":"2026090809073964100_cit4","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijpharm.2020.119560","article-title":"Amorphous\nSolid Dispersions: An Update for Preparation, Characterization, Mechanism\non Bioavailability, Stability, Regulatory Considerations and Marketed\nProducts","volume":"586","author":"Pandi","year":"2020","journal-title":"Int. J. Pharm."},{"key":"2026090809073964100_cit5","doi-asserted-by":"publisher","DOI":"10.3389\/fddev.2023.1232012","article-title":"Lipid-Based\nOral Formulation in Capsules to Improve the Delivery of Poorly Water-Soluble\nDrugs","volume":"3","author":"Mohite","year":"2023","journal-title":"Front. Drug Delivery"},{"key":"2026090809073964100_cit6","doi-asserted-by":"publisher","first-page":"1125","DOI":"10.1007\/s11095-021-03063-3","article-title":"Lipophilic Salts\nand Lipid-Based Formulations: Enhancing the Oral Delivery of Octreotide","volume":"38","author":"Li","year":"2021","journal-title":"Pharm. Res."},{"key":"2026090809073964100_cit7","doi-asserted-by":"publisher","first-page":"1307","DOI":"10.1208\/s12249-014-0156-1","article-title":"Preclinical Formulations: Insight,\nStrategies, and\nPractical Considerations","volume":"15","author":"Shah","year":"2014","journal-title":"AAPS PharmSciTech"},{"key":"2026090809073964100_cit8","doi-asserted-by":"publisher","first-page":"1089","DOI":"10.3390\/ph18081089","article-title":"Novel Strategies for\nthe Formulation of Poorly Water-Soluble\nDrug Substances by Different Physical Modification Strategies with\na Focus on Peroral Applications","volume":"18","author":"Quodbach","year":"2025","journal-title":"Pharmaceuticals"},{"key":"2026090809073964100_cit9","doi-asserted-by":"publisher","first-page":"1099","DOI":"10.3390\/life13051099","article-title":"Advancement in Solubilization\nApproaches: A Step towards Bioavailability Enhancement of Poorly Soluble\nDrugs","volume":"13","author":"Kumari","year":"2023","journal-title":"Life"},{"key":"2026090809073964100_cit10","doi-asserted-by":"publisher","first-page":"3054","DOI":"10.1016\/j.xphs.2024.07.017","article-title":"Oral Absorption from Surfactant-Based Drug Formulations:\nThe Impact of Molecularly Dissolved Drug on Bioavailability","volume":"113","author":"Holzem","year":"2024","journal-title":"J. Pharm. Sci."},{"key":"2026090809073964100_cit11","doi-asserted-by":"publisher","first-page":"288","DOI":"10.3390\/pharmaceutics17030288","article-title":"Cyclodextrins: Enhancing Drug Delivery,\nSolubility and Bioavailability for Modern Therapeutics","volume":"17","author":"Nicolaescu","year":"2025","journal-title":"Pharmaceutics"},{"key":"2026090809073964100_cit12","doi-asserted-by":"publisher","first-page":"603","DOI":"10.1016\/j.addr.2007.05.010","article-title":"Salt\nFormation to Improve Drug Solubility","volume":"59","author":"Serajuddin","year":"2007","journal-title":"Adv. Drug\nDelivery Rev."},{"key":"2026090809073964100_cit13","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1016\/j.ejpb.2022.04.002","article-title":"Combining\nEnabling Formulation Strategies\nto Generate Supersaturated Solutions of Delamanid: In Situ Salt Formation\nduring Amorphous Solid Dispersion Fabrication for More Robust Release\nProfiles","volume":"174","author":"Duong","year":"2022","journal-title":"Eur. J. Pharm. Biopharm."},{"key":"2026090809073964100_cit14","doi-asserted-by":"publisher","first-page":"1300","DOI":"10.3109\/03639045.2012.727829","article-title":"Amorphous\nSolid Dispersion Successfully Improved Oral Exposure of ADX71943 in\nSupport of Toxicology Studies","volume":"39","author":"Ayad","year":"2013","journal-title":"Drug Dev. Ind.\nPharm."},{"key":"2026090809073964100_cit15","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1016\/j.ijpharm.2013.04.037","article-title":"Development of Amorphous\nSolid Dispersion Formulations of a Poorly\nWater-Soluble Drug, MK-0364","volume":"452","author":"Sotthivirat","year":"2013","journal-title":"Int. J. Pharm."},{"key":"2026090809073964100_cit16","doi-asserted-by":"publisher","first-page":"16","DOI":"10.3390\/colloids7010016","article-title":"Recent Advances\nin Improving the\nBioavailability of Hydrophobic\/Lipophilic Drugs and Their Delivery\nvia Self-Emulsifying Formulations","volume":"7","author":"Ameta","year":"2023","journal-title":"Colloids\nInterfaces"},{"key":"2026090809073964100_cit17","doi-asserted-by":"publisher","DOI":"10.3389\/fvets.2022.938021","article-title":"Case Report:\nSuccessful\nIntravenous Lipid Emulsion Therapy for Canine Amphetamine Toxicosis","volume":"9","author":"Harris","year":"2022","journal-title":"Front. Vet. Sci."},{"key":"2026090809073964100_cit18","doi-asserted-by":"publisher","first-page":"413","DOI":"10.1023\/a:1016212804288","article-title":"A Theoretical Basis\nfor a Biopharmaceutic Drug Classification: The Correlation of in Vitro\nDrug Product Dissolution and in Vivo Bioavailability","volume":"12","author":"Amidon","year":"1995","journal-title":"Pharm. Res."},{"key":"2026090809073964100_cit19","doi-asserted-by":"publisher","first-page":"1031","DOI":"10.1021\/mp500504q","article-title":"Preclinical Dose Number and Its Application\nin Understanding\nDrug Absorption Risk and Formulation Design for Preclinical Species","volume":"12","author":"Wuelfing","year":"2015","journal-title":"Mol. Pharmaceutics"},{"key":"2026090809073964100_cit20","doi-asserted-by":"publisher","first-page":"1685","DOI":"10.1021\/acs.jmedchem.1c01687","article-title":"Dose Number as a Tool to Guide Lead\nOptimization for Orally Bioavailable Compounds in Drug Discovery","volume":"65","author":"Wuelfing","year":"2022","journal-title":"J. Med. Chem."},{"key":"2026090809073964100_cit21","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1002\/psp4.13313","article-title":"Comparing\nScientific Machine Learning With Population Pharmacokinetic and Classical\nMachine Learning Approaches for Prediction of Drug Concentrations","volume":"14","author":"Valderrama","year":"2025","journal-title":"CPT: Pharmacomet. Syst. Pharmacol."},{"key":"2026090809073964100_cit22","doi-asserted-by":"publisher","DOI":"10.3389\/fphar.2024.1330855","article-title":"Predicting\nPharmacodynamic Effects through Early Drug Discovery with Artificial\nIntelligence-Physiologically Based Pharmacokinetic (AI-PBPK) Modelling","volume":"15","author":"Wu","year":"2024","journal-title":"Front. Pharmacol."},{"key":"2026090809073964100_cit23","doi-asserted-by":"publisher","DOI":"10.1111\/cts.13824","article-title":"Machine\nLearning Framework to Predict Pharmacokinetic Profile of Small Molecule\nDrugs Based on Chemical Structure","volume":"17","author":"Pillai","year":"2024","journal-title":"Clin. Transl.\nSci."},{"key":"2026090809073964100_cit24","doi-asserted-by":"publisher","first-page":"1530","DOI":"10.3390\/pharmaceutics14081530","article-title":"Machine Learning and Pharmacometrics for Prediction\nof Pharmacokinetic Data: Differences, Similarities and Challenges\nIllustrated with Rifampicin","volume":"14","author":"Keutzer","year":"2022","journal-title":"Pharmaceutics"},{"key":"2026090809073964100_cit25","doi-asserted-by":"publisher","first-page":"2976","DOI":"10.1021\/acs.molpharmaceut.4c01431","article-title":"Machine Learning Prediction\nand Validation\nof Plasma Concentration\u2013Time Profiles","volume":"22","author":"Iwata","year":"2025","journal-title":"Mol. Pharmaceutics"},{"key":"2026090809073964100_cit26","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/a:1010933404324","article-title":"Random Forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"2026090809073964100_cit27","unstructured":"Chen, T.; Guestrin, C.\n          XGBoost-\nA Scalable\nTree Boosting System.Pdf. In  Proceedings of\nthe 22nd ACM SIGKDD International Conference on Knowledge Discovery\nand Data Mining, San Francisco, 13\u201317 201610.1145\/2939672.2939785."},{"key":"2026090809073964100_cit28","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/ade568","article-title":"Investigating the Benefits of Artificial\nNeural Networks over Linear Approaches to BMI Decoding","volume":"22","author":"Temmar","year":"2025","journal-title":"J. Neural Eng."},{"key":"2026090809073964100_cit29","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1038\/s41586-024-08328-6","article-title":"Accurate Predictions\non Small Data with a Tabular Foundation\nModel","volume":"637","author":"Hollmann","year":"2025","journal-title":"Nature"},{"key":"2026090809073964100_cit30","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1039\/c7sc02664a","article-title":"MoleculeNet: A Benchmark\nfor Molecular Machine Learning","volume":"9","author":"Wu","year":"2018","journal-title":"Chem. Sci."},{"key":"2026090809073964100_cit31","doi-asserted-by":"crossref","unstructured":"Beyazit, E.; Kozaczuk, J.; Li, B.; Wallace, V.; Fadlallah, B.\n          An Inductive Bias for Tabular\nDeep Learning. In  37th Conference on Neural\nInformation Processing Systems, 2023.","DOI":"10.52202\/075280-1869"},{"key":"2026090809073964100_cit32","doi-asserted-by":"publisher","first-page":"296","DOI":"10.1016\/j.ijpharm.2016.06.048","article-title":"In Vitro and\nin Vivo Investigation of the Gastrointestinal Behavior of Simvastatin","volume":"510","author":"Geboers","year":"2016","journal-title":"Int. J. Pharm."},{"key":"2026090809073964100_cit33","doi-asserted-by":"publisher","first-page":"5217","DOI":"10.1021\/jm070705c","article-title":"1,3-Disubstituted\nUreas Functionalized with Ether Groups\nAre Potent Inhibitors of the Soluble Epoxide Hydrolase with Improved\nPharmacokinetic Properties","volume":"50","author":"Kim","year":"2007","journal-title":"J. Med. Chem."},{"key":"2026090809073964100_cit34","doi-asserted-by":"publisher","first-page":"7757","DOI":"10.3390\/molecules28237757","article-title":"Urea-Functionalized Heterocycles:\nStructure, Hydrogen Bonding and Applications","volume":"28","author":"Keszei","year":"2023","journal-title":"Molecules"},{"key":"2026090809073964100_cit35","doi-asserted-by":"publisher","first-page":"719","DOI":"10.3390\/molecules21060719","article-title":"Hydrogen Bonding: Between\nStrengthening the Crystal Packing and Improving Solubility of Three\nHaloperidol Derivatives","volume":"21","author":"Saluja","year":"2016","journal-title":"Molecules"},{"key":"2026090809073964100_cit36","doi-asserted-by":"publisher","first-page":"501","DOI":"10.1208\/s12249-008-9071-7","article-title":"Effect of Drug Lipophilicity\nand Ionization on Permeability\nAcross the Buccal Mucosa: A Technical Note","volume":"9","author":"Kokate","year":"2008","journal-title":"AAPS PharmSciTech"}],"container-title":["ACS Omega"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acsomega.6c05839","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/35\/53186\/67413016\/acsomega.6c05839.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/35\/53186\/67413016\/acsomega.6c05839.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,8]],"date-time":"2026-09-08T13:07:45Z","timestamp":1788872865000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/acsodf\/article\/11\/35\/53186\/5329334\/Machine-Learning-Driven-Prediction-of-Dose-Linear"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,26]]},"references-count":36,"journal-issue":{"issue":"35","published-online":{"date-parts":[[2026,8,26]]},"published-print":{"date-parts":[[2026,9,8]]}},"URL":"https:\/\/doi.org\/10.1021\/acsomega.6c05839","relation":{},"ISSN":["2470-1343"],"issn-type":[{"value":"2470-1343","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,26]]}}}