EUVIMEDEUVIMED entdecken
Open Access

CryptoBank: A resource for the identification and prediction of cryptic sites in proteins

Pedro Febrer Martinez iD, Thorben Fröhlking iD, Alberto Borsatto iD, Francesco L Gervasio iD

Science Advances · 2026

DOI: 10.1126/sciadv.ady6364 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, 2026. Verfasst von Pedro Febrer Martinez und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Pharmaceutical Sciences, University of Geneva, Geneva 1206, Switzerland. Hauptthemen laut MEDLINE: Proteins, Databases, Protein, Computational Biology. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 62 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: EPSRC Centre for Doctoral Training in Medical Imaging, PRACE and the Swiss National Supercomputing Centre, Swiss National Science Foun-dation (SNSF) and BRIDGE.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cryptic binding sites offer opportunities to modulate targets previously considered “undruggable.” However, the scarcity of validated examples limits the development of predictive tools. Here, we introduce CryptoBank, a large-scale database of cryptic sites identified by applying a machine learning model to detect ligand-induced conformational changes in more than 6 million structural alignments of unbound (apo) and bound (holo) protein pairs from the Protein Data Bank (PDB). Our analysis reveals that cryptic pockets are widespread, occurring in ~18% of protein clusters. Leveraging this resource, we fine-tuned a protein language model (PLM) to predict cryptic sites directly from protein sequences. Crucially, we show the broad applicability of our strategy by predicting cryptic sites in four proteins with low sequence identity to any CryptoBank entry and then validating these predictions using molecular dynamics simulations. CryptoBank and the predictive PLM are publicly accessible via a web server, providing valuable resources for cryptic site discovery.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsDatabases, ProteinComputational BiologyBinding SitesMolecular Dynamics SimulationLigandsProtein ConformationProtein BindingMachine Learning

Substanzen

ProteinsLigands

Autor:innen und Institutionen

  • School of Pharmaceutical Sciences, University of Geneva, Geneva 1206, Switzerland.
  • Institute of Pharmaceutical Sciences of Western Switzerland, University of Geneva, Geneva 1206, Switzerland.
  • Swiss Bioinformatics Institute, University of Geneva, Geneva 1206, Switzerland.
  • Chemistry Department, University College London (UCL), London WC1E 6BT, UK.

ORCID-Profile: Pedro Febrer Martinez, Thorben Fröhlking, Alberto Borsatto, Francesco L Gervasio

Förderung

  • EPSRC Centre for Doctoral Training in Medical Imaging
  • PRACE and the Swiss National Supercomputing Centre
  • Swiss National Science Foun-dation (SNSF) and BRIDGE
  • BBSRC and UC PhD CASE studentship

Zitieren

APA

Martinez, P. F., Fröhlking, T., Borsatto, A., & Gervasio, F. L. (2026). CryptoBank: A resource for the identification and prediction of cryptic sites in proteins. Science Advances. https://doi.org/10.1126/sciadv.ady6364

Vancouver

Martinez PF, Fröhlking T, Borsatto A, Gervasio FL. CryptoBank: A resource for the identification and prediction of cryptic sites in proteins. Science Advances. 2026. doi:10.1126/sciadv.ady6364

BibTeX
@article{Martinez2026,
  author = {Martinez, Pedro Febrer and Fröhlking, Thorben and Borsatto, Alberto and Gervasio, Francesco L},
  title = {CryptoBank: A resource for the identification and prediction of cryptic sites in proteins},
  journal = {Science Advances},
  year = {2026},
  doi = {10.1126/sciadv.ady6364}
}
Kostenlos starten →