GATSBI: improving context-aware protein embeddings through biologically motivated data splits
Bioinformatics · 01.07.2026 · 42 (Supplement_1)
DOI: 10.1093/bioinformatics/btag233 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 07.07.2026. Verfasst von Gowri Nayar und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Data Science, Stanford University, CA, 94304, United States. Hauptthemen laut MEDLINE: Proteins, Computational Biology, Software, Protein Interaction Mapping. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NLM NIH HHS, NIH HHS, Burroughs Welcome Fund.
Abstract (Originalsprache)
Motivation Understanding protein function requires integrating diverse biological evidence while accounting for strong contextual dependence. Recent protein embedding methods increasingly leverage heterogeneous biological networks, yet their evaluation protocols often fail to reflect the specific biological tasks for which the embeddings are intended. Prediction of missing interactions, annotation of new proteins, and discovery of functional modules require fundamentally different data partitions, such as edge-masked versus node-held-out splits. Moreover, most approaches report performance primarily on well-studied proteins, where computational predictions are least needed, risking substantial overestimation of real-world utility. Results We introduce a graph attention-based framework (GATSBI) to construct context-aware protein embeddings from integrated protein–protein interactions, co-expression, sequence representations, and tissue-specific associations. Using task-aligned evaluation protocols, we show that models trained with biologically appropriate partitions achieve markedly better generalization. Across interaction, function, and functional set prediction, Gatsbi consistently outperforms existing pretrained embeddings for both well-studied and understudied proteins, with the largest gains observed for the understudied regime and under inductive node-held-out evaluation. To enable broad reuse, we provide the learned embeddings for download for application to other protein prediction tasks. Availability and implementation https://github.com/Helix-Research-Lab/GATSBI-embedding.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Biomedical Data Science, Stanford University, CA, 94304, United States.
- Department of Bioengineering, Stanford University, CA, 94304, United States.
- Department of Genetics, Stanford University, CA, 94304, United States.
- Department of Medicine, Stanford University, CA, 94304, United States.
ORCID-Profile: Gowri Nayar, Russ B Altman
Förderung
- NLM NIH HHS
- NIH HHS
- Burroughs Welcome Fund
- NLM
- NIH
Zitieren
APA
Nayar, G., & Altman, R. B. (2026). GATSBI: improving context-aware protein embeddings through biologically motivated data splits. Bioinformatics, 42(Supplement_1). https://doi.org/10.1093/bioinformatics/btag233
Vancouver
Nayar G, Altman RB. GATSBI: improving context-aware protein embeddings through biologically motivated data splits. Bioinformatics. 2026;42(Supplement_1). doi:10.1093/bioinformatics/btag233
BibTeX
@article{Nayar2026,
author = {Nayar, Gowri and Altman, Russ B},
title = {GATSBI: improving context-aware protein embeddings through biologically motivated data splits},
journal = {Bioinformatics},
year = {2026},
volume = {42},
number = {Supplement_1},
doi = {10.1093/bioinformatics/btag233}
}
EUVIMED