Machine learning–driven discovery of antimicrobial peptides against Pseudomonas aeruginosa
Frontiers in Pharmacology · 24.06.2026 · 17
DOI: 10.3389/fphar.2026.1837055 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Pharmacology, online erschienen am 24.06.2026. Verfasst von YaTong Zhang und 8 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Medicine, School of Pharmacy, Macau University of Science and Technology, Taipa, Macau, China. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Antibiotic resistance has become a global health crisis, driving the urgent need for novel antibacterial strategies. Antimicrobial peptides (AMPs) have emerged as promising therapeutic candidates due to their broad-spectrum activity and low propensity for resistance development. However, conventional discovery approaches remain time-consuming and inefficient for large-scale screening.
Methods
In this study, we developed an integrated prediction framework that combines deep learning with classical machine learning methods to enable rapid and accurate identification of AMPs. The model was applied to screen over 250,000 artificially generated peptide sequences. Candidate peptides were selected based on physicochemical descriptors and activity-associated features, followed by solid-phase synthesis and minimum inhibitory concentration assays against Pseudomonas aeruginosa. Mechanistic investigations were conducted using scanning electron microscopy to visualize membrane damage, molecular dynamics simulations to probe peptide-membrane interactions, and transcriptomic profiling to assess bacterial stress responses and metabolic pathway alterations.
Results
Ten promising antibacterial candidates were successfully identified and experimentally validated. All selected peptides exhibited measurable activity against P. aeruginosa, with several showing potent inhibitory effects. Microscopic and simulation analyses confirmed that the peptides exert their antibacterial effects primarily through membrane disruption. Transcriptomic data further revealed that these peptides interfere with key metabolic pathways and activate bacterial stress response systems.
Discussion
Our findings demonstrate that the integrated deep learning-machine learning pipeline offers an efficient and reliable approach for large-scale AMP discovery. The mechanistic insights gained from this study not only validate the predicted candidates but also provide a foundation for rational optimization of peptide-based therapeutics. This comprehensive strategy holds promise for accelerating the development of next-generation alternatives to conventional antibiotics.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Faculty of Medicine, School of Pharmacy, Macau University of Science and Technology, Taipa, Macau, China.
- Department of Pharmacy, Beijing Hospital, National Center of Gerontology; Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, China.
- Department of Pharmacy, Peking University Third Hospital, Beijing, China.
- Institute for Drug Evaluation, Peking University Health Science Center, Beijing, China.
- Department of Pharmacy, Beijing Friendship Hospital, Capital Medical University, Beijing, China.
- College of Agroforestry and Medicine, The Open University of China, Beijing, China.
Zitieren
APA
Zhang, Y., Sun, X., Li, H., Li, D., Yu, C., Zhao, B., Zhou, Y., Zhu, Y. Z., & Zhao, R. (2026). Machine learning–driven discovery of antimicrobial peptides against Pseudomonas aeruginosa. Frontiers in Pharmacology, 17. https://doi.org/10.3389/fphar.2026.1837055
Vancouver
Zhang Y, Sun X, Li H, Li D, Yu C, Zhao B, et al. Machine learning–driven discovery of antimicrobial peptides against Pseudomonas aeruginosa. Frontiers in Pharmacology. 2026;17. doi:10.3389/fphar.2026.1837055
BibTeX
@article{Zhang2026,
author = {Zhang, YaTong and Sun, Xuelin and Li, HuiBo and Li, DanDan and Yu, Caiyuan and Zhao, Bin and Zhou, YingChen and Zhu, Yi Zhun and Zhao, Rongsheng},
title = {Machine learning–driven discovery of antimicrobial peptides against Pseudomonas aeruginosa},
journal = {Frontiers in Pharmacology},
year = {2026},
volume = {17},
doi = {10.3389/fphar.2026.1837055}
}
EUVIMED