EUVIMEDEUVIMED entdecken
Crossref

DynamicGT: A dynamic-aware geometric transformer model to predict protein-binding interfaces in flexible and disordered regions

Omid Mokhtari, Sergei Grudinin, Yasaman Karami, Hamed Khakzad

Cell Systems · 2026

DOI: 10.1016/j.cels.2025.101454 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cell Systems, 2026. Verfasst von Omid Mokhtari und 3 weiteren Autor:innen. Beteiligte Einrichtung: Université de Lorraine, CNRS, Inria, LORIA, 54000 Nancy, France. Hauptthemen laut MEDLINE: Proteins. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 74 Literaturangaben. Gefördert von: Grand Équipement National De Calcul Intensif, French National Research Agency.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Protein-protein interactions are fundamental to cellular processes, yet current deep learning approaches for binding site prediction rely on static structures, limiting their accuracy for disordered or flexible regions. We introduce dynamic geometric transformer (DynamicGT), a dynamic-aware model that integrates conformational dynamics into a cooperative graph neural network (Co-GNN) with a GT. Our model encodes dynamic features at both node (atom) and edge (interaction) levels, considering bound and unbound states to improve generalization. Dynamic regulation of messages passing between core and surface residues enhances detection of critical interactions for efficient information flow. Trained on a 1-ms molecular dynamics simulation dataset and augmented with AlphaFlow-generated conformations, the model was benchmarked extensively. Evaluation on diverse datasets containing disordered, transient, and unbound structures demonstrates that incorporating dynamics within a cooperative architecture significantly improves prediction accuracy where flexibility is key while requiring substantially less data than leading static approaches.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsProtein BindingBinding SitesMolecular Dynamics SimulationNeural Networks, ComputerProtein ConformationComputational BiologyHumansAlgorithmsDeep Learning

Stichwörter der Autor:innen

cooperative message passingdynamic-aware systemsgeometric transformersintrinsically disordered proteinsprotein-binding site prediction

Substanzen

Proteins

Autor:innen und Institutionen

  • Université de Lorraine, CNRS, Inria, LORIA, 54000 Nancy, France.
  • CNRS, Grenoble INP, LJK, Université Grenoble Alpes, Grenoble, France.
  • Université de Lorraine, CNRS, Inria, LORIA, 54000 Nancy, France. Electronic address: yasaman.karami@inria.fr.
  • Université de Lorraine, CNRS, Inria, LORIA, 54000 Nancy, France. Electronic address: hamed.khakzad@inria.fr.

Förderung

  • Grand Équipement National De Calcul Intensif
  • French National Research Agency

Zitieren

APA

Mokhtari, O., Grudinin, S., Karami, Y., & Khakzad, H. (2026). DynamicGT: A dynamic-aware geometric transformer model to predict protein-binding interfaces in flexible and disordered regions. Cell Systems. https://doi.org/10.1016/j.cels.2025.101454

Vancouver

Mokhtari O, Grudinin S, Karami Y, Khakzad H. DynamicGT: A dynamic-aware geometric transformer model to predict protein-binding interfaces in flexible and disordered regions. Cell Systems. 2026. doi:10.1016/j.cels.2025.101454

BibTeX
@article{Mokhtari2026,
  author = {Mokhtari, Omid and Grudinin, Sergei and Karami, Yasaman and Khakzad, Hamed},
  title = {DynamicGT: A dynamic-aware geometric transformer model to predict protein-binding interfaces in flexible and disordered regions},
  journal = {Cell Systems},
  year = {2026},
  doi = {10.1016/j.cels.2025.101454}
}
Kostenlos starten →