EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Zero-Shot Protein–Ligand Binding Site Prediction from Protein Sequence and SMILES

Mahdi Pourmirzaei, Salhuldin Alqarghuli, Kai Chen, Mohammadreza Pourmirzaei, Dong Xu iD

Bioinformatics · 12.11.2026

DOI: 10.1093/bioinformatics/btag826 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 12.11.2026. Verfasst von Mahdi Pourmirzaei und 4 weiteren Autor:innen. Beteiligte Einrichtung: Electrical Engineering and Computer Science, University of Missouri, Columbia, Missouri, USA. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Accurate identification of protein–ligand binding sites is critical for mechanistic biology and drug discovery, yet predictive performance varies substantially across ligand families and data-availability regimes. We present a systematic ligand-aware benchmark and evaluation framework for residue-level binding-site prediction across three settings: overrepresented ligands with abundant examples, underrepresented ligands with limited examples, and zero-shot ligands unseen during training. Results Using a BioLiP2-derived dataset with ligand-specific sequence clustering, we evaluate a progressive three-stage sequence-based modeling suite designed to study these regimes. Stage 1 trains ligand-specific predictors using a pretrained protein language model (PLM). Stage 2 introduces ligand conditioning through learned embeddings, enabling a single multi-ligand model. Stage 3 replaces these embeddings with representations from a pretrained chemical language model (CLM) operating on SMILES, enabling prediction for previously unseen ligands. In the original three-stage comparison, Stage 2 achieved a higher ligand-macro F1 than Stage 1 on the overrepresented test set (0.5832 versus 0.4769), although the comparison simultaneously changes predictor architecture, ligand conditioning, and the training regime. Stage 2 also outperformed several sequence- and structure-based baselines. Stage 3 achieves a pooled residue-level F1 of 0.3109 and a ligand-macro F1 of 0.2338 on previously unseen ligands, while remaining competitive on represented ligands. Control experiments using constant, mismatched, and scrambled SMILES inputs reduce performance across all three test regimes, providing evidence that the model uses information associated with the supplied ligand input rather than relying solely on generic protein binding-site propensity. Ablations across five PLM scales and multiple CLMs further show that increasing PLM scale generally improves performance, although the trend is not strictly monotonic, whereas CLM scaling produces smaller and less consistent gains. Overall, this study provides a systematic benchmark for quantifying the effect of ligand data availability on residue-level binding-site prediction and for evaluating generalization to previously unseen ligands. Availability and Implementation The source code and model weights are available at https://github.com/mahdip72/ProteinLigand. The datasets are available at https://doi.org/10.5281/zenodo.23042604. Supplementary Information Supplementary data are available at Bioinformatics online.

Schlagwörter

Stichwörter der Autor:innen

Protein language modelbinding site predictionchemical language modelsequence-based binding-site predictionzero-shot prediction

Autor:innen und Institutionen

  • Electrical Engineering and Computer Science, University of Missouri, Columbia, Missouri, USA.
  • College of Biomedical Informatics, University of Information Technology and Communications, Baghdad, Iraq.
  • Department of Architecture, Politecnico di Milano, Milan, Italy.
  • Health Informatics Institute, Morsani College of Medicine, University of South Florida, Tampa, Florida, USA.

ORCID-Profile: Dong Xu

Zitieren

APA

Pourmirzaei, M., Alqarghuli, S., Chen, K., Pourmirzaei, M., & Xu, D. (2026). Zero-Shot Protein–Ligand Binding Site Prediction from Protein Sequence and SMILES. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag826

Vancouver

Pourmirzaei M, Alqarghuli S, Chen K, Pourmirzaei M, Xu D. Zero-Shot Protein–Ligand Binding Site Prediction from Protein Sequence and SMILES. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag826

BibTeX
@article{Pourmirzaei2026,
  author = {Pourmirzaei, Mahdi and Alqarghuli, Salhuldin and Chen, Kai and Pourmirzaei, Mohammadreza and Xu, Dong},
  title = {Zero-Shot Protein–Ligand Binding Site Prediction from Protein Sequence and SMILES},
  journal = {Bioinformatics},
  year = {2026},
  doi = {10.1093/bioinformatics/btag826}
}
Kostenlos starten →