EUVIMEDEUVIMED entdecken
Crossref

Predicting Subcutaneous Antibody Bioavailability Using Ensemble Protein Language Models

Miles Cabreza, William Hojegian, I-En Wu, Pin-Kuang Lai iD

Molecular Pharmaceutics · 01.09.2025 · 22 (9) · 5504-5511

DOI: 10.1021/acs.molpharmaceut.5c00523 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Molecular Pharmaceutics, online erschienen am 05.08.2025. Verfasst von Miles Cabreza und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemical Engineering and Materials Science, Stevens Institute of Technology, Hoboken 07030, New Jersey United States. Hauptthemen laut MEDLINE: Antibodies, Monoclonal. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 33 Literaturangaben. Gefördert von: University of Delaware.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Monoclonal antibodies are pivotal in modern therapeutics, yet predicting their subcutaneous bioavailability remains challenging due to the intricacies of the SC environment and the limitations of traditional experimental models. In this study, we introduce a novel machine learning framework that leverages protein language models (PLMs) to derive high-dimensional embeddings directly from antibody sequences. Using three distinct PLMs─antiBERTy, ABlang, and ESM-2─we extracted numerical representations that were subsequently refined via feature selection and dimensionality reduction. A systematic evaluation of multiple classifiers using Leave-One-Out cross-validation led us to develop a robust ensemble model based on a tuned support vector machine classifier, which achieved a validation accuracy of 89%. This ensemble approach, which aggregates predictions across antibodies, outperforms prior computational methods. To facilitate broad accessibility, we deployed the model as a web application, SubQAvail, enabling rapid bioavailability predictions from input antibody sequences. Our findings demonstrate the potential of integrating PLM-derived features with ensemble learning to enhance the predictive accuracy and scalability of mAb bioavailability assessment, thereby accelerating the therapeutic development pipeline.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Antibodies, MonoclonalBiological AvailabilityMachine LearningHumansSupport Vector MachineInjections, SubcutaneousAnimals

Stichwörter der Autor:innen

bioavailabilityhigh-concentration antibody formulationmachine learningprotein language modelssubcutaneous injection

Substanzen

Antibodies, Monoclonal

Autor:innen und Institutionen

  • Department of Chemical Engineering and Materials Science, Stevens Institute of Technology, Hoboken 07030, New Jersey United States.

ORCID-Profile: Pin-Kuang Lai

Förderung

  • University of Delaware

Zitieren

APA

Cabreza, M., Hojegian, W., Wu, I. E., & Lai, P. K. (2025). Predicting Subcutaneous Antibody Bioavailability Using Ensemble Protein Language Models. Molecular Pharmaceutics, 22(9), 5504-5511. https://doi.org/10.1021/acs.molpharmaceut.5c00523

Vancouver

Cabreza M, Hojegian W, Wu IE, Lai PK. Predicting Subcutaneous Antibody Bioavailability Using Ensemble Protein Language Models. Molecular Pharmaceutics. 2025;22(9):5504-5511. doi:10.1021/acs.molpharmaceut.5c00523

BibTeX
@article{Cabreza2025,
  author = {Cabreza, Miles and Hojegian, William and Wu, I-En and Lai, Pin-Kuang},
  title = {Predicting Subcutaneous Antibody Bioavailability Using Ensemble Protein Language Models},
  journal = {Molecular Pharmaceutics},
  year = {2025},
  volume = {22},
  number = {9},
  pages = {5504--5511},
  doi = {10.1021/acs.molpharmaceut.5c00523}
}
Kostenlos starten →