EUVIMEDOpdag EUVIMED
VergleichsstudieOpen access · CC BY-NC 4.0

PCANN Program for Structure‐Based Prediction of Protein–Protein Binding Affinity: Comparison With Other Neural‐Network Predictors

Olga O Lebedenko, Mikhail S Polovinkin, Anastasiia A Kazovskaia, Nikolai R Skrynnikov iD

Proteins: Structure, Function, and Bioinformatics · 2025

DOI: 10.1002/prot.26821 ↗

At a glance

Vergleichsstudie in Proteins: Structure, Function, and Bioinformatics, 2025. Written by Olga O Lebedenko and 3 co-authors. Institution: Laboratory of Biomolecular NMR, St. Petersburg State University, St. Petersburg, Russia. Major MEDLINE topics: Neural Networks, Computer, Proteins, Software, Computational Biology. Cited 1 times so far (Crossref); 43 references. The full text is freely available (open access, CC BY-NC 4.0). Funded by: Saint Petersburg State University.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

In this communication, we introduce a new structure‐based affinity predictor for protein–protein complexes. This predictor, dubbed PCANN (Protein Complex Affinity by Neural Network), uses the ESM‐2 language model to encode the information about protein binding interfaces and graph attention network (GAT) to parlay this information into predictions. In the tests employing two previously unused literature‐extracted datasets, PCANN performed better than the best of the publicly available predictors, BindPPI, with mean absolute error (MAE) of 1.3 versus 1.4 kcal/mol. Further progress in the development of predictors using deep learning models is faced with two problems: (i) the amount of experimental data available to train and test new predictors is limited and (ii) the available data are often not very accurate and lack internal consistency with respect to measurement conditions. These issues can be potentially addressed through an AI‐leveraged literature search followed by careful human curation and by introducing additional parameters to account for variations in experimental conditions.

Subjects

MeSH terms (MEDLINE)

Neural Networks, ComputerProteinsSoftwareComputational BiologyProtein BindingHumansBinding SitesDatabases, ProteinDeep LearningProtein ConformationAlgorithms

Author keywords

ESM‐2 language modelKd prediction programcomparison of Kd predictorsdeep learninggraph attention networkprotein binding databasesprotein–protein binding

Substances

Proteins

Authors and institutions

  • Laboratory of Biomolecular NMR, St. Petersburg State University, St. Petersburg, Russia.
  • Faculty of Mathematics & Computer Science, St. Petersburg State University, St. Petersburg, Russia.
  • Department of Chemistry, Purdue University, West Lafayette, Indiana, USA.

ORCID profiles: Nikolai R Skrynnikov

Funding

  • Saint Petersburg State University

Cite

APA

Lebedenko, O. O., Polovinkin, M. S., Kazovskaia, A. A., & Skrynnikov, N. R. (2025). PCANN Program for Structure‐Based Prediction of Protein–Protein Binding Affinity: Comparison With Other Neural‐Network Predictors. Proteins: Structure, Function, and Bioinformatics. https://doi.org/10.1002/prot.26821

Vancouver

Lebedenko OO, Polovinkin MS, Kazovskaia AA, Skrynnikov NR. PCANN Program for Structure‐Based Prediction of Protein–Protein Binding Affinity: Comparison With Other Neural‐Network Predictors. Proteins: Structure, Function, and Bioinformatics. 2025. doi:10.1002/prot.26821

BibTeX
@article{Lebedenko2025,
  author = {Lebedenko, Olga O and Polovinkin, Mikhail S and Kazovskaia, Anastasiia A and Skrynnikov, Nikolai R},
  title = {PCANN Program for Structure‐Based Prediction of Protein–Protein Binding Affinity: Comparison With Other Neural‐Network Predictors},
  journal = {Proteins: Structure, Function, and Bioinformatics},
  year = {2025},
  doi = {10.1002/prot.26821}
}
Start for free →