EUVIMEDEUVIMED entdecken
Open Access

KRstereo: Predicting β-Hydroxy Stereochemistry in Polyketides Using Protein Language Models

Hsin-Ying Tsai, Wenqiang Xu iD, Wen Jun Xie iD, Yousong Ding iD

Journal of Chemical Information and Modeling · 2026

DOI: 10.1021/acs.jcim.6c02706 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, 2026. Verfasst von Hsin-Ying Tsai und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medicinal Chemistry and Center for Natural Products, Drug Discovery and Development (CNPD3), University of Florida, Gainesville, Florid…. Hauptthemen laut MEDLINE: Polyketides, Polyketide Synthases, Machine Learning. Die Arbeit nennt 71 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, University of Florida, National Institute of General Medical Sciences.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Polyketides are a major class of bioactive natural products whose activities are often determined by the stereochemistry of β-hydroxyl groups. In type I polyketide synthases (PKSs), ketoreductase (KR) domains establish these stereocenters, making accurate prediction of KR stereochemistry important for natural product discovery and PKS engineering. Existing rule-based methods rely on a limited set of sequence motifs and often perform poorly across phylogenetically diverse taxa. Here, we present KRstereo, a machine learning framework that predicts KR stereochemistry directly from sequence. Analysis of β-modular KR domains from MIBiG 3.1 identified informative sequence features beyond canonical motifs, motivating the use of protein language model embeddings. KRstereo achieved accuracies of up to 95.7% across taxa and 93.0% for non-Streptomyces KRs, consistently outperforming existing rule-based approaches. Validation using newly characterized KR domains from MIBiG 4.0 confirmed strong generalizability, including cases misclassified by current methods. Application of KRstereo to 20,840 β-modular KR domains from antiSMASH enabled large-scale stereochemical annotation of previously uncharacterized PKS systems. By linking sequence to stereochemical function, KRstereo improves reconstruction of polyketide structures from biosynthetic gene clusters and facilitates stereochemistry-aware genome mining and engineering of PKS assembly lines.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PolyketidesPolyketide SynthasesMachine LearningStereoisomerismModels, Molecular

Substanzen

PolyketidesPolyketide Synthases

Autor:innen und Institutionen

  • Department of Medicinal Chemistry and Center for Natural Products, Drug Discovery and Development (CNPD3), University of Florida, Gainesville, Florida32610, United States.

ORCID-Profile: Wenqiang Xu, Wen Jun Xie, Yousong Ding

Förderung

  • NIGMS NIH HHS
  • University of Florida
  • National Institute of General Medical Sciences

Zitieren

APA

Tsai, H. Y., Xu, W., Xie, W. J., & Ding, Y. (2026). KRstereo: Predicting β-Hydroxy Stereochemistry in Polyketides Using Protein Language Models. Journal of Chemical Information and Modeling. https://doi.org/10.1021/acs.jcim.6c02706

Vancouver

Tsai HY, Xu W, Xie WJ, Ding Y. KRstereo: Predicting β-Hydroxy Stereochemistry in Polyketides Using Protein Language Models. Journal of Chemical Information and Modeling. 2026. doi:10.1021/acs.jcim.6c02706

BibTeX
@article{Tsai2026,
  author = {Tsai, Hsin-Ying and Xu, Wenqiang and Xie, Wen Jun and Ding, Yousong},
  title = {KRstereo: Predicting β-Hydroxy Stereochemistry in Polyketides Using Protein Language Models},
  journal = {Journal of Chemical Information and Modeling},
  year = {2026},
  doi = {10.1021/acs.jcim.6c02706}
}
Kostenlos starten →