EUVIMEDEUVIMED entdecken
Crossref

Bilingua‐DPP‐IV: A Bimodal Fusion Convolutional Neural Network Model for Prediction of Dipeptidyl Peptidase IV Inhibitory Peptides

Zhixing Zhang, Qiule Yu, Chen Yang, Hao Duan, Jiaojiao Fang, Jiahao Xu, Changda Gong, Weihua Li iD, Guixia Liu iD, Yun Tang iD

ChemMedChem · 28.10.2026 · 21 (20)

DOI: 10.1002/cmdc.70551 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ChemMedChem, online erschienen am 23.10.2026. Verfasst von Zhixing Zhang und 9 weiteren Autor:innen. Beteiligte Einrichtung: Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East…. Hauptthemen laut MEDLINE: Dipeptidyl-Peptidase IV Inhibitors, Dipeptidyl Peptidase 4, Peptides, Neural Networks, Computer. Die Arbeit nennt 32 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation: Dipeptidyl peptidase IV (DPP‐IV) is a key therapeutic target for type 2 diabetes, and its inhibitory peptides are a promising class of therapeutic agents. However, the discovery of these promising drugs has been restricted, as many existing computational predictors are based on datasets that have remained static for several years. Results: Here, we first presented a newly curated and expanded benchmark dataset, then introduced Bilingua‐DPP‐IV, a bimodal deep learning framework that integrates features from both amino acid sequences and molecular structures. Our approach leveraged embeddings from the large‐scale pretrained models ESM C and Uni‐Mol2, processed through a convolutional neural network with an optimized gated fusion mechanism. On an independent test set, the Bilingua‐DPP‐IV ensemble model achieved a Matthews correlation coefficient of 0.764 and an area under the receiver operating characteristic curve of 0.963. On an external validation set, the model also demonstrated an accuracy of 0.898, exceeding the performance of existing state‐of‐the‐art predictors, such as StructuralDPPIV.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Dipeptidyl-Peptidase IV InhibitorsDipeptidyl Peptidase 4PeptidesNeural Networks, ComputerConvolutional Neural NetworksHumansPrediction AlgorithmsDeep Learning

Stichwörter der Autor:innen

artificial intelligencecomputer scienceconvolutional neural networkdeep learningensemble forecastingensemble learningsource codetest setweb server

Substanzen

Dipeptidyl-Peptidase IV InhibitorsDipeptidyl Peptidase 4Peptides

Autor:innen und Institutionen

  • Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East China University of Science and Technology, Shanghai, China.

ORCID-Profile: Weihua Li, Guixia Liu, Yun Tang

Förderung

  • National Natural Science Foundation of China
  • Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism

Zitieren

APA

Zhang, Z., Yu, Q., Yang, C., Duan, H., Fang, J., Xu, J., Gong, C., Li, W., Liu, G., & Tang, Y. (2026). Bilingua‐DPP‐IV: A Bimodal Fusion Convolutional Neural Network Model for Prediction of Dipeptidyl Peptidase IV Inhibitory Peptides. ChemMedChem, 21(20). https://doi.org/10.1002/cmdc.70551

Vancouver

Zhang Z, Yu Q, Yang C, Duan H, Fang J, Xu J, et al. Bilingua‐DPP‐IV: A Bimodal Fusion Convolutional Neural Network Model for Prediction of Dipeptidyl Peptidase IV Inhibitory Peptides. ChemMedChem. 2026;21(20). doi:10.1002/cmdc.70551

BibTeX
@article{Zhang2026,
  author = {Zhang, Zhixing and Yu, Qiule and Yang, Chen and Duan, Hao and Fang, Jiaojiao and Xu, Jiahao and Gong, Changda and Li, Weihua and Liu, Guixia and Tang, Yun},
  title = {Bilingua‐DPP‐IV: A Bimodal Fusion Convolutional Neural Network Model for Prediction of Dipeptidyl Peptidase IV Inhibitory Peptides},
  journal = {ChemMedChem},
  year = {2026},
  volume = {21},
  number = {20},
  doi = {10.1002/cmdc.70551}
}
Kostenlos starten →