The role of protein embeddings for protein–protein interaction prediction with graph neural networks
Briefings in Bioinformatics · 2026
DOI: 10.1093/bib/bbag313 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Laura Balbi und 3 weiteren Autor:innen. Beteiligte Einrichtung: LASIGE, Faculdade de Ciências da Universidade de Lisboa, Campo Grande, 1749-016 Lisbon, Portugal. Hauptthemen laut MEDLINE: Computational Biology, Graph Neural Networks, Protein Interaction Mapping, Protein Interaction Maps. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 208 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: LASIGE Research Unit, European Union's Horizon 2020 research and innovation programme, CancerScan project by the EU's HORIZON Europe research and innovation programme.
Abstract (Originalsprache)
Graph neural networks are a natural fit for protein–protein interaction (PPI) prediction because they exploit the graph structure of interactomes, yet it remains unclear how much they benefit from protein-level representations and which sources of protein information matter most. We address this by surveying and organizing the field into a coherent taxonomy that describes three paradigms: network-level approaches (shallow and deep graph representation learning), protein-level approaches (sequence, structure, and function), and hybrid methods that combine both paradigms—providing a cohesive synthesis that highlights their respective strengths, complementarities, and tradeoffs. We developed a unified framework to systematically compare these paradigms that incorporates multiple representative methods and evaluates them on two widely used benchmarks: OGBl-ppa and HuRI. Our results indicate that hybrid methods have improved performance, with function-driven representations emerging as the most informative protein-level representations for PPI prediction. This work provides a taxonomical map of the field, a reproducible framework for evaluating representative methods across paradigms, and practical guidance for future PPI prediction efforts.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- LASIGE, Faculdade de Ciências da Universidade de Lisboa, Campo Grande, 1749-016 Lisbon, Portugal.
- University of Mannheim, 68131 Mannheim, Germany.
- Department of Computer Science, City St George's, University of London, EC1V 0HB, United Kingdom.
Förderung
- LASIGE Research Unit
- European Union's Horizon 2020 research and innovation programme
- CancerScan project by the EU's HORIZON Europe research and innovation programme
- European Union’s Horizon 2020 research and innovation programme
- CancerScan project by the EU’s HORIZON Europe research and innovation programme
Zitieren
APA
Balbi, L., Sousa, R. T., Cotovio, P., & Pesquita, C. (2026). The role of protein embeddings for protein–protein interaction prediction with graph neural networks. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag313
Vancouver
Balbi L, Sousa RT, Cotovio P, Pesquita C. The role of protein embeddings for protein–protein interaction prediction with graph neural networks. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag313
BibTeX
@article{Balbi2026,
author = {Balbi, Laura and Sousa, Rita T and Cotovio, Pedro and Pesquita, Catia},
title = {The role of protein embeddings for protein–protein interaction prediction with graph neural networks},
journal = {Briefings in Bioinformatics},
year = {2026},
doi = {10.1093/bib/bbag313}
}
EUVIMED