EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

AlphaFold model quality self‐assessment improvement via deep graph learning

Jacob Verburgt, Zicong Zhang, Daisuke Kihara iD

Protein Science · 2025-09 · 34 (9)

DOI: 10.1002/pro.70274 ↗

At a glance

Journal article in Protein Science, published online on 2025-08-18. Written by Jacob Verburgt and 2 co-authors. Institution: Department of Biological Sciences, Purdue University, Indiana, USA. Major MEDLINE topics: Deep Learning, Proteins, Protein Folding, Models, Molecular. Cited 9 times so far (Crossref); 31 references. The full text is freely available (open access, CC BY 4.0). Funded by: NIH HHS, National Science Foundation, NIGMS NIH HHS.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

In recent years, significant advancements have been made in deep learning‐based computational modeling of proteins, with DeepMind's AlphaFold2 standing out as a landmark achievement. These computationally modeled protein structures not only provide atomic coordinates but also include self‐confidence metrics to assess the relative quality of the modeling, either for individual residues or the entire protein. However, these self‐confidence scores are not always reliable; for instance, poorly modeled regions of a protein may sometimes be assigned high confidence. To address this limitation, we introduce Equivariant Quality Assessment Folding (EQAFold), an enhanced framework that refines the Local Distance Difference Test prediction head of AlphaFold to generate more accurate self‐confidence scores. Our results demonstrate that EQAFold outperforms the standard AlphaFold architecture and recent model quality assessment protocols in providing more reliable confidence metrics. Source code for EQAFold is available at https://github.com/kiharalab/EQAFold_public .

Subjects

MeSH terms (MEDLINE)

Deep LearningProteinsProtein FoldingModels, MolecularSoftwareProtein Conformation

Author keywords

AlphaFolddeep learningmodel quality assessmentprotein structure prediction

Substances

Proteins

Authors and institutions

  • Department of Biological Sciences, Purdue University, Indiana, USA.
  • Rosen Center for Advanced Computing, Purdue University, Indiana, USA.
  • Department of Computer Science, Purdue University, Indiana, USA.

ORCID profiles: Daisuke Kihara

Funding

  • NIH HHS
  • National Science Foundation
  • NIGMS NIH HHS
  • National Institute of General Medical Sciences
  • National Institutes of Health

Cite

APA

Verburgt, J., Zhang, Z., & Kihara, D. (2025). AlphaFold model quality self‐assessment improvement via deep graph learning. Protein Science, 34(9). https://doi.org/10.1002/pro.70274

Vancouver

Verburgt J, Zhang Z, Kihara D. AlphaFold model quality self‐assessment improvement via deep graph learning. Protein Science. 2025;34(9). doi:10.1002/pro.70274

BibTeX
@article{Verburgt2025,
  author = {Verburgt, Jacob and Zhang, Zicong and Kihara, Daisuke},
  title = {AlphaFold model quality self‐assessment improvement via deep graph learning},
  journal = {Protein Science},
  year = {2025},
  volume = {34},
  number = {9},
  doi = {10.1002/pro.70274}
}
Start for free →