EUVIMEDEUVIMED entdecken
Crossref

ProtGeoNet-Pocket: A Binding Site Prediction Approach Integrating Sequence, Geometry, and Graph Structure

Mingjian Jiang iD, Zhi Zhang iD, Teng Ma iD, Huaibin Hang, Yaping Fan, Shunpeng Pang iD, Wei Zhou, Yuanyuan Zhang

Journal of Chemical Information and Modeling · 13.10.2025 · 65 (19) · 10736-10753

DOI: 10.1021/acs.jcim.5c01568 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, online erschienen am 17.09.2025. Verfasst von Mingjian Jiang und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Information and Control Engineering, Qingdao University of Technology, No. 777 Jialingjiang East Road, Huangdao District, Qingdao, Shandong…. Hauptthemen laut MEDLINE: Proteins. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 54 Literaturangaben. Gefördert von: Department of Science and Technology of Shandong Province, Scientific Research Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

ProtGeoNet-Pocket is an innovative multimodal prediction framework designed for protein binding site recognition, effectively integrating sequence information, geometric features, and graph-based structural representations. To address the structural complexity of proteins and the diversity of binding pocket shapes, ProtGeoNet-Pocket leverages multiscale structural information for high-precision binding site prediction. It uses a PointNet module to extract geometric features from residue coordinates and enhances them using an attention mechanism. The geometric features are then fused with encoded sequence features and graph edge features. The combined features are fed into a Graph Isomorphism Network (GIN) to capture topological relationships via a message-passing mechanism. ProtGeoNet-Pocket achieved an F1 score of 72.87% on the scPDB training set and demonstrated strong predictive performance across five independent benchmark data sets: COACH420, HOLO4K, SC6K, PDBbind, and ApoHolo. Furthermore, the visualization results confirm a high spatial overlap between the predicted and actual binding sites, demonstrating the superior performance of this method compared to existing ones.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsBinding SitesProtein ConformationModels, Molecular

Substanzen

Proteins

Autor:innen und Institutionen

  • School of Information and Control Engineering, Qingdao University of Technology, No. 777 Jialingjiang East Road, Huangdao District, Qingdao, Shandong 266520, China.
  • School of Computer Engineering, Weifang University, No. 5147 Dongfeng East Street, Kuiwen District, Weifang, Shandong 261061, China.

ORCID-Profile: Mingjian Jiang, Zhi Zhang, Teng Ma, Shunpeng Pang

Förderung

  • Department of Science and Technology of Shandong Province
  • Scientific Research Foundation

Zitieren

APA

Jiang, M., Zhang, Z., Ma, T., Hang, H., Fan, Y., Pang, S., Zhou, W., & Zhang, Y. (2025). ProtGeoNet-Pocket: A Binding Site Prediction Approach Integrating Sequence, Geometry, and Graph Structure. Journal of Chemical Information and Modeling, 65(19), 10736-10753. https://doi.org/10.1021/acs.jcim.5c01568

Vancouver

Jiang M, Zhang Z, Ma T, Hang H, Fan Y, Pang S, et al. ProtGeoNet-Pocket: A Binding Site Prediction Approach Integrating Sequence, Geometry, and Graph Structure. Journal of Chemical Information and Modeling. 2025;65(19):10736-10753. doi:10.1021/acs.jcim.5c01568

BibTeX
@article{Jiang2025,
  author = {Jiang, Mingjian and Zhang, Zhi and Ma, Teng and Hang, Huaibin and Fan, Yaping and Pang, Shunpeng and Zhou, Wei and Zhang, Yuanyuan},
  title = {ProtGeoNet-Pocket: A Binding Site Prediction Approach Integrating Sequence, Geometry, and Graph Structure},
  journal = {Journal of Chemical Information and Modeling},
  year = {2025},
  volume = {65},
  number = {19},
  pages = {10736--10753},
  doi = {10.1021/acs.jcim.5c01568}
}
Kostenlos starten →