Developing SCL2205 : a protein sequence-based spatial modelling dataset for the protein language model frontier
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag666 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Daniel Ouso und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science, University College Dublin - Belfield, Dublin D04 N2E5, Ireland. Hauptthemen laut MEDLINE: Proteins, Deep Learning, Computational Biology, Sequence Analysis, Protein. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Research Ireland, Research Training in Genomics Data Science.
Abstract (Originalsprache)
Motivation Deep learning (DL) has substantially advanced protein subcellular localization (SCL) prediction, yet its potential remains constrained by suboptimal input preparation and limited high-quality reference data. Furthermore, existing state-of-the-art (SoTA) predictors suffer from performance metric inflation due to unmitigated training-to-testing data leakage during homology augmentation. We address these challenges by introducing SCL2205, a leak-minimized benchmark dataset and pipeline curated specifically to support trustworthy, scalable, and reproducible DL-based SCL modelling. Results SCL2205 was constructed from the universal protein knowledgebase (UniProtKB) using rigorous preprocessing, manual label mapping, and stringent partitioning. When evaluated on independent test sets, SCL2205 yielded up to a 10.8 percentage point improvement in macro area under the precision–recall curve (PR-AUC) over SoTA baselines (mean Δ 95% CI = 0.07–0.12), with maximum benefits observed when paired with modern protein language models (PLMs). Crucially, we quantify for the first time a systemic 5.2%±0.32 data leakage rate in conventional homology augmentation workflows—even when restricting sequence similarity searches to just 10% of the training set. Availability and implementation The dataset is openly available on Dryad under a CC0 1.0 Universal licence (https://doi.org/10.5061/dryad.2ngf1vj1t). The dataset interface is available as an installable Python package, p-scldata (v2026.2.0), under the MIT licence on the Python Package Index (PyPI). Full code and data repositories are hosted on GitHub (https://github.com/ousodaniel/scldata) and archived on Zenodo (https://doi.org/10.5281/zenodo.21796423).
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- School of Computer Science, University College Dublin - Belfield, Dublin D04 N2E5, Ireland.
- Centre for Research Training in Genomics Data Science, University of Galway, Galway H91 TK33, Ireland.
ORCID-Profile: Daniel Ouso, Gianluca Pollastri
Förderung
- Research Ireland
- Research Training in Genomics Data Science
Zitieren
APA
Ouso, D., & Pollastri, G. (2026). Developing SCL2205 : a protein sequence-based spatial modelling dataset for the protein language model frontier. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag666
Vancouver
Ouso D, Pollastri G. Developing SCL2205 : a protein sequence-based spatial modelling dataset for the protein language model frontier. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag666
BibTeX
@article{Ouso2026,
author = {Ouso, Daniel and Pollastri, Gianluca},
title = {Developing SCL2205 : a protein sequence-based spatial modelling dataset for the protein language model frontier},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag666}
}
EUVIMED