EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Modeling intrinsically disordered regions from AlphaFold2 to AlphaFold3

Mahta Mehdiabadi iD, Silvio C E Tosatto iD, Damiano Piovesan iD

Protein Science · 2026-01 · 35 (1)

DOI: 10.1002/pro.70426 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Protein Science, online erschienen am 27.12.2025. Verfasst von Mahta Mehdiabadi und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Sciences, University of Padova, Padua, Italy. Hauptthemen laut MEDLINE: Intrinsically Disordered Proteins, Models, Molecular, Software. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NextGenerationEU, HORIZON EUROPE Marie Sklodowska-Curie Actions, Ministero dell'Istruzione, dell'Università e della Ricerca.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

AlphaFold2 has demonstrated a remarkable success in predicting the structures of globular proteins and folded domains with near‐experimental accuracy. However, it typically represents intrinsically disordered regions (IDRs), protein segments that lack a stable 3D structure under physiological conditions, as long extended loops that appear to float around the structured core. While AlphaFold2's static prediction cannot capture the conformational heterogeneity and the dynamic nature of IDRs, it performs well in predicting IDRs from sequence. AlphaFold3 introduces significant architectural and training modifications over its predecessor, including the use of cross‐distillation aimed at reducing structural hallucinations in disordered regions. In this study, we look into how these models differ in representing IDRs. We evaluate the performance of AlphaFold3 and AlphaFold2 on disorder prediction, using the CAID3 benchmark. Our analysis shows that AlphaFold3 does not outperform AlphaFold2 in this benchmark. We observe that solvent accessibility remains a robust and consistent proxy for predicting intrinsic disorder across both models. However, changes in the predicted secondary structure content and pLDDT scores lead to different interpretations of disorder. Overall, our findings suggest that AlphaFold2 remains the preferred choice for identifying intrinsically disordered regions, as it avoids structural hallucinations while providing predictions comparable to those of AlphaFold3.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Intrinsically Disordered ProteinsModels, MolecularSoftwareProtein FoldingProtein ConformationProtein Structure, Secondary

Stichwörter der Autor:innen

AlphaFoldcritical assessment of protein intrinsic disorderhallucinationsintrisic protein disorderstructure prediction

Substanzen

Intrinsically Disordered Proteins

Autor:innen und Institutionen

  • Department of Biomedical Sciences, University of Padova, Padua, Italy.
  • Institute of Biomembranes, Bioenergetics and Molecular Biotechnologies, National Research Council (CNR-IBIOM), Bari, Italy.

ORCID-Profile: Mahta Mehdiabadi, Silvio C E Tosatto, Damiano Piovesan

Förderung

  • NextGenerationEU
  • HORIZON EUROPE Marie Sklodowska-Curie Actions
  • Ministero dell'Istruzione, dell'Università e della Ricerca

Zitieren

APA

Mehdiabadi, M., Tosatto, S. C. E., & Piovesan, D. (2026). Modeling intrinsically disordered regions from AlphaFold2 to AlphaFold3. Protein Science, 35(1). https://doi.org/10.1002/pro.70426

Vancouver

Mehdiabadi M, Tosatto SCE, Piovesan D. Modeling intrinsically disordered regions from AlphaFold2 to AlphaFold3. Protein Science. 2026;35(1). doi:10.1002/pro.70426

BibTeX
@article{Mehdiabadi2026,
  author = {Mehdiabadi, Mahta and Tosatto, Silvio C E and Piovesan, Damiano},
  title = {Modeling intrinsically disordered regions from AlphaFold2 to AlphaFold3},
  journal = {Protein Science},
  year = {2026},
  volume = {35},
  number = {1},
  doi = {10.1002/pro.70426}
}
Kostenlos starten →