EUVIMEDEUVIMED entdecken
Crossref

Context-informed subgraph foundation models enable interpretable protein-function prediction

Zhuomin Zhou, Jiahua Rao, Zhongyue Zhang, Runze Ma, Jiancheng Yang, Shuangjia Zheng

Cell Systems · 2026

DOI: 10.1016/j.cels.2026.101535 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cell Systems, 2026. Verfasst von Zhuomin Zhou und 5 weiteren Autor:innen. Beteiligte Einrichtung: Global Institute of Future Technology, Shanghai Jiao Tong University, Shanghai, China. Hauptthemen laut MEDLINE: Proteins, Computational Biology. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 64 Literaturangaben. Gefördert von: Science and Technology Commission of Shanghai Municipality, National Natural Science Foundation of China, Shanghai Municipal Natural Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Protein-function prediction is crucial for elucidating molecular mechanisms driving biological processes and therapeutics development. Despite numerous computational tools demonstrating promising performance, they fall short when predicting rare, uncharacterized functions or indirect activities. Here, we present COSMOS, a context-aware Gene Ontology (GO) subgraph mining system for protein-function prediction. By leveraging inductive subgraph foundation models and an enriched knowledge graph of protein-GO relationships, COSMOS performs zero-shot, few-shot, and low-homology protein-function prediction. Built on 7,923,952 functional semantic relationships, COSMOS demonstrates robust capabilities to (1) generate state-of-the-art predictions for GO classes with sparse or no experimental annotations, (2) provide interpretable functional subgraphs for transparent rationale analysis, and (3) deliver complementary benefits when integrated with existing embedding-based prediction methods. We anticipate that COSMOS will serve as a complementary approach to conventional protein annotation methods and an interpretable tool for predicting protein functions within underexplored GO classes, thereby advancing genomics and therapeutic research.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsComputational BiologyGene OntologyHumansData MiningPrediction AlgorithmsDatabases, ProteinMolecular Sequence AnnotationAlgorithms

Stichwörter der Autor:innen

Gene Ontologycomputational modelsknowledge graphprotein-function predictions

Substanzen

Proteins

Autor:innen und Institutionen

  • Global Institute of Future Technology, Shanghai Jiao Tong University, Shanghai, China.
  • School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China.
  • Computer Vision Laboratory, Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland.
  • Global Institute of Future Technology, Shanghai Jiao Tong University, Shanghai, China. Electronic address: shuangjia.zheng@sjtu.edu.cn.

Förderung

  • Science and Technology Commission of Shanghai Municipality
  • National Natural Science Foundation of China
  • Shanghai Municipal Natural Science Foundation

Zitieren

APA

Zhou, Z., Rao, J., Zhang, Z., Ma, R., Yang, J., & Zheng, S. (2026). Context-informed subgraph foundation models enable interpretable protein-function prediction. Cell Systems. https://doi.org/10.1016/j.cels.2026.101535

Vancouver

Zhou Z, Rao J, Zhang Z, Ma R, Yang J, Zheng S. Context-informed subgraph foundation models enable interpretable protein-function prediction. Cell Systems. 2026. doi:10.1016/j.cels.2026.101535

BibTeX
@article{Zhou2026,
  author = {Zhou, Zhuomin and Rao, Jiahua and Zhang, Zhongyue and Ma, Runze and Yang, Jiancheng and Zheng, Shuangjia},
  title = {Context-informed subgraph foundation models enable interpretable protein-function prediction},
  journal = {Cell Systems},
  year = {2026},
  doi = {10.1016/j.cels.2026.101535}
}
Kostenlos starten →