EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physical Implausibility of Carbohydrate Ligands in Results of Deep Learning-Based Cofolding Methods

Muhammad Luthfi, Adam J Simpkin, Luc G Elliott, Pornthep Sompornpisut iD, Daniel J Rigden iD

Journal of Chemical Information and Modeling · 2026

DOI: 10.1021/acs.jcim.5c03075 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, 2026. Verfasst von Muhammad Luthfi und 4 weiteren Autor:innen. Beteiligte Einrichtung: Program in Bioinformatics and Computational Biology, College of Interdisciplinary and Integrative Studies, Chulalongkorn University, Bangkok 10330, T…. Hauptthemen laut MEDLINE: Carbohydrates, Deep Learning. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Chulalongkorn University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Stereochemistry violations in AlphaFold 3 models are more prevalent than currently appreciated. Analysis of 900 carbohydrate ligands revealed that 85.8% have errors, mainly in chirality but also including bond conversions (15.2%), planar ring distortions (3.9%), aromatic ring formations (2.5%), and improper structural configurations (0.9%). Boltz-1x reduced most violations dramatically but increased improper configurations to 22.1%, notably in N-acetyl-α-neuraminic acid. The BondedAtomPairs protocol reduced stereochemical issues but lost the reducing-end anomeric oxygen, highlighting ongoing challenges in accurate carbohydrate modeling.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

CarbohydratesDeep LearningLigandsStereoisomerismModels, MolecularCarbohydrate Conformation

Substanzen

LigandsCarbohydrates

Autor:innen und Institutionen

  • Program in Bioinformatics and Computational Biology, College of Interdisciplinary and Integrative Studies, Chulalongkorn University, Bangkok 10330, Thailand.
  • Department of Biochemistry, Cell and Systems Biology; Institute of Systems, Molecular and Integrative Biology, University of Liverpool, Crown Street, Liverpool L69 7ZB, United Kingdom.
  • Center of Excellence in Computational Chemistry, Department of Chemistry, Chulalongkorn University, Bangkok 10330, Thailand.

ORCID-Profile: Pornthep Sompornpisut, Daniel J Rigden

Förderung

  • Chulalongkorn University

Zitieren

APA

Luthfi, M., Simpkin, A. J., Elliott, L. G., Sompornpisut, P., & Rigden, D. J. (2026). Physical Implausibility of Carbohydrate Ligands in Results of Deep Learning-Based Cofolding Methods. Journal of Chemical Information and Modeling. https://doi.org/10.1021/acs.jcim.5c03075

Vancouver

Luthfi M, Simpkin AJ, Elliott LG, Sompornpisut P, Rigden DJ. Physical Implausibility of Carbohydrate Ligands in Results of Deep Learning-Based Cofolding Methods. Journal of Chemical Information and Modeling. 2026. doi:10.1021/acs.jcim.5c03075

BibTeX
@article{Luthfi2026,
  author = {Luthfi, Muhammad and Simpkin, Adam J and Elliott, Luc G and Sompornpisut, Pornthep and Rigden, Daniel J},
  title = {Physical Implausibility of Carbohydrate Ligands in Results of Deep Learning-Based Cofolding Methods},
  journal = {Journal of Chemical Information and Modeling},
  year = {2026},
  doi = {10.1021/acs.jcim.5c03075}
}
Kostenlos starten →