Comprehensive evaluation of AlphaFold/OpenFold prediction of experimentally unresolved proteins through novel metrics
Briefings in Bioinformatics · 2026
DOI: 10.1093/bib/bbag461 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Florencia R Díaz und 7 weiteren Autor:innen. Beteiligte Einrichtung: Facultad de Ciencias Exactas Físicas y Naturales (FCEFyN), Universidad Nacional de Córdoba, Av. Vélez Sarsfield 1611, Ciudad Universitaria, X5016GCA,…. Hauptthemen laut MEDLINE: Proteins, Computational Biology, Protein Folding, Software. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Predicting accurate protein structures is essential for understanding molecular mechanisms, interpreting the impact of sequence variation, and supporting translational applications ranging from drug discovery to clinical genomics. Recent advances in deep-learning–based predictors such as AlphaFold2, OpenFold, and AlphaFold3 have transformed structural biology, enabling routine in silico modeling even for challenging or previously uncharacterized proteins. However, systematic benchmarking of these tools—especially for novel targets and single amino acid variants—remains limited. Conventional global metrics often fail to capture biologically meaningful discrepancies. By evaluating multiple implementations of AlphaFold2 and OpenFold, together with ColabFold and the AlphaFold3 server, across 10 different proteins and 222 single amino acid protein variants encompassing a wide range of sizes, structures, and functions, we show that although widely used global indicators—like mean pLDDT, pTM-score, and RMSD—frequently suggest comparable performance, substantial local-level differences remain elusive. To address this gap, we introduce a comparative framework leveraging Bland–Altman agreement analysis, to evaluate per-residue Cα-confidence differences and Per-Residue profiles (PRPs), complemented by Uniform Manifold Approximation and Projection (UMAP). This approach reveals marked localized divergences, particularly within flexible or intrinsically disordered regions, where both predictor choice and single-residue substitutions trigger the largest conformational shifts. We further demonstrate that using reduced homology databases has minimal impact on predicted structural quality, offering computationally efficient alternatives. Collectively, our findings underscore the importance of integrating global and residue-specific evaluations to more accurately assess robustness, agreement, and practical usability across contemporary protein structure prediction methods.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Facultad de Ciencias Exactas Físicas y Naturales (FCEFyN), Universidad Nacional de Córdoba, Av. Vélez Sarsfield 1611, Ciudad Universitaria, X5016GCA, Córdoba, Argentina.
- ScireLab, Fundación para el Progreso de la Medicina, 9 de Julio 941, Centro, X5000EMS, Córdoba, Argentina.
- Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Córdoba, Argentina.
- Facultad de Ingeniería, Universidad Católica de Córdoba, Av. Armada Argentina 3555, Campus Universitario, X5016DHK, Córdoba, Argentina.
- Grupo GNOCIX, Facultad de Ciencias e Ingenierías, Universidad del Sinú Elías Bechara Zainum, Cra. 1W No. 38-153, Barrio Juan XXIII, 230001, Montería, Córdoba, Colombia.
- Departamento de Bioquímica Clínica, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, ica, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, Haya de la Torre y Medina Allende s/n, Ciudad Univ…
ORCID-Profile: Elmer A Fernández
Zitieren
APA
Díaz, F. R., Orschanski, D., Folco, J. I., Ceci, J. E., Nibeyro, G., Robles Vega, H., Nicola, J. P., & Fernández, E. A. (2026). Comprehensive evaluation of AlphaFold/OpenFold prediction of experimentally unresolved proteins through novel metrics. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag461
Vancouver
Díaz FR, Orschanski D, Folco JI, Ceci JE, Nibeyro G, Robles Vega H, et al. Comprehensive evaluation of AlphaFold/OpenFold prediction of experimentally unresolved proteins through novel metrics. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag461
BibTeX
@article{Daz2026,
author = {Díaz, Florencia R and Orschanski, Daniela and Folco, Juan I and Ceci, Juana Espain and Nibeyro, Guadalupe and Robles Vega, Horderlin and Nicola, Juan P and Fernández, Elmer A},
title = {Comprehensive evaluation of AlphaFold/OpenFold prediction of experimentally unresolved proteins through novel metrics},
journal = {Briefings in Bioinformatics},
year = {2026},
doi = {10.1093/bib/bbag461}
}
EUVIMED