EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

DeepPANB: integrating protein language model with PaiNN equivariant graph neural networks for prediction of protein–nucleic acid binding sites

Jilong Zhang, Zhixiang Wu, Jingjie Su, Xinyu Zhang, Yue Li, Zihan Li, Chunhua Li iD

Briefings in Bioinformatics · 2026

DOI: 10.1093/bib/bbag450 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Jilong Zhang und 6 weiteren Autor:innen. Beteiligte Einrichtung: College of Chemistry and Life Science, Beijing University of Technology, Pingleyuan 100, ChaoYang District, Beijing 100124, China. Hauptthemen laut MEDLINE: Computational Biology, Proteins, Deep Learning, DNA. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Protein–nucleic acid interactions are fundamental to various biological processes, and accurately identifying nucleic acid binding sites on proteins is essential for understanding gene regulation mechanisms and advancing drug design. Here, we present DeepPANB, an effective deep learning model for the issue. DeepPANB is built upon the Polarizable Atom Interaction Neural Network (PaiNN) architecture, an E(3)-equivariant graph neural network, which explicitly couples scalar and vector channels through lightweight message passing, thereby enabling effective modeling of the direction- and distance-dependent residue interactions. DeepPANB integrates multiple feature types including sequence embeddings from the Ankh pretrained model, residue–nucleotide pairwise propensity extracted by us from the large-scale datasets, as well as residue intrinsic disorder, physicochemical properties, and structural descriptors. To our best knowledge, PaiNN and Ankh are first introduced here for protein–nucleic acid binding site prediction. On the protein–DNA test set, DeepPANB achieves state-of-the-art performance, outperforming the existing methods. On protein–RNA test set, DeepPANB, fine-tuned with the feature types unchanged, shows competitive performance. Overall, DeepPANB demonstrates robust and generalizable performance, providing an effective tool for protein–nucleic acid binding site prediction. Freely available at https://github.com/ChunhuaLab/DeepPANB.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Computational BiologyProteinsDeep LearningDNANucleic AcidsSoftwareRNAGraph Neural NetworksBinding SitesProtein BindingAlgorithms

Stichwörter der Autor:innen

binding site predictionequivariant graph neural networkprotein language modelprotein–nucleic acid interactions

Substanzen

ProteinsDNANucleic AcidsRNA

Autor:innen und Institutionen

  • College of Chemistry and Life Science, Beijing University of Technology, Pingleyuan 100, ChaoYang District, Beijing 100124, China.

ORCID-Profile: Chunhua Li

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Zhang, J., Wu, Z., Su, J., Zhang, X., Li, Y., Li, Z., & Li, C. (2026). DeepPANB: integrating protein language model with PaiNN equivariant graph neural networks for prediction of protein–nucleic acid binding sites. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag450

Vancouver

Zhang J, Wu Z, Su J, Zhang X, Li Y, Li Z, et al. DeepPANB: integrating protein language model with PaiNN equivariant graph neural networks for prediction of protein–nucleic acid binding sites. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag450

BibTeX
@article{Zhang2026,
  author = {Zhang, Jilong and Wu, Zhixiang and Su, Jingjie and Zhang, Xinyu and Li, Yue and Li, Zihan and Li, Chunhua},
  title = {DeepPANB: integrating protein language model with PaiNN equivariant graph neural networks for prediction of protein–nucleic acid binding sites},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  doi = {10.1093/bib/bbag450}
}
Kostenlos starten →