EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MoEPH: an adaptive fusion-based LLM for predicting phage-host interactions in health informatics

Qian Chen, Zihang Zhao, Min Li, Wenchen Song, Minfeng Xiao, Min Fang

Frontiers in Microbiology · 18.09.2025 · 16

DOI: 10.3389/fmicb.2025.1634705 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Microbiology, online erschienen am 18.09.2025. Verfasst von Qian Chen und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Artificial Intelligence and National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University, Shenzhen, Guangd…. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Phage-host interaction prediction plays a crucial role in the development of phage therapy, particularly in combating antimicrobial resistance (AMR). Current in silico models often suffer from limited generalizability and low interpretability. To address these gaps, we introduce MoEPH (Mixture-of-Experts for Phage-Host prediction), a novel framework that integrates transformer-based protein embeddings (ProtBERT and ProT5) with domain-specific statistical descriptors. Our model dynamically combines features using a gated fusion mechanism, ensuring robust and adaptive prediction. We evaluate MoEPH on three publicly available phage-host interaction databases: Dataset 1 (101 host strains, 129 phages), Dataset 2 (38 host strains, 176 phages), and Dataset 3 (combined). Experimental results demonstrate that MoEPH outperforms existing methods, achieving an accuracy of 99.6% on balanced datasets and a 31% improvement on highly imbalanced data. The model provides a transparent, dynamic and knowledge-driven fusion solution for phage-host prediction, contributing to more effective phage therapy recommendations. Future work will focus on incorporating structural protein features and exploring alternative neural backbones for further enhancement.

Schlagwörter

Stichwörter der Autor:innen

interpretablemixture of experts (MoE)robustnesstransformer-based protein embeddingstrustworthy phage-host prediction

Autor:innen und Institutionen

  • School of Artificial Intelligence and National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University, Shenzhen, Guangdong, China.
  • School of Computer Science and Technology, The University of Hong Kong, Hong Kong, Hong Kong SAR, China.
  • BGI-Shenzhen, Shenzhen, China.
  • University of Chinese Academy of Sciences, Beijing, China.
  • Education Center of Experiments and Innovations, Harbin Institute of Technology, Shenzhen, Guangdong, China.

Zitieren

APA

Chen, Q., Zhao, Z., Li, M., Song, W., Xiao, M., & Fang, M. (2025). MoEPH: an adaptive fusion-based LLM for predicting phage-host interactions in health informatics. Frontiers in Microbiology, 16. https://doi.org/10.3389/fmicb.2025.1634705

Vancouver

Chen Q, Zhao Z, Li M, Song W, Xiao M, Fang M. MoEPH: an adaptive fusion-based LLM for predicting phage-host interactions in health informatics. Frontiers in Microbiology. 2025;16. doi:10.3389/fmicb.2025.1634705

BibTeX
@article{Chen2025,
  author = {Chen, Qian and Zhao, Zihang and Li, Min and Song, Wenchen and Xiao, Minfeng and Fang, Min},
  title = {MoEPH: an adaptive fusion-based LLM for predicting phage-host interactions in health informatics},
  journal = {Frontiers in Microbiology},
  year = {2025},
  volume = {16},
  doi = {10.3389/fmicb.2025.1634705}
}
Kostenlos starten →