Machine learning modeling of zebrafish toxicity endpoints after exposure to PROTACs
Toxicological Sciences · 30.01.2026 · 209 (1)
DOI: 10.1093/toxsci/kfaf162 ↗Kurz gefasst
Fachartikel in der Zeitschrift Toxicological Sciences, online erschienen am 19.11.2025. Verfasst von Christopher Yogodzinski und 9 weiteren Autor:innen. Beteiligte Einrichtung: Collaborations Pharmaceuticals, Inc, Raleigh, NC 27606, United States. Hauptthemen laut MEDLINE: Zebrafish, Machine Learning, Toxicity Tests. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 65 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIEHS NIH HHS, National Institute of Environmental Health Sciences (NIEHS), NIGMS NIH HHS.
Abstract (Originalsprache)
Zebrafish (Danio rerio) are an ideal system for understanding developmental toxicity as they display similar toxicity outcomes to other vertebrates. Further, many molecules have been tested for developmental toxicity in zebrafish, providing an opportunity for machine learning model development. We curated 1,345 small molecules from ToxCast, flame retardant compounds, per- and polyfluoroalkyl substances (PFAS), and industrial chemicals published by the Superfund Research Program (SRP). Following curation, we trained machine learning models on the zebrafish toxicity endpoints ANY_ = any effect including mortality, ANY_BUT_MORT = any effect excluding mortality, MORT = mortality, i.e. did the embryo die, EDEM = did an edema form, and CRAN = Craniofacial malformation. We demonstrated that these models were better than random when compared with shuffled data. We also fine-tuned the molecular SMILES encoder MolBART to predict on all zebrafish toxicity endpoints and found it generally matched the performance of classical machine learning models for ANY_BUT_MORT, CRAN, and EDEM endpoints. We present new toxicity data for Proteolysis Targeting Chimeras (PROTACs) in zebrafish and machine learning models for these data by fingerprinting different parts of the molecule individually, yielding predictive performance (AUROC 0.6 to 0.7). If we are to reduce animal testing with new approach methodologies like these zebrafish toxicity models they need to be able adapt to new molecular classes like PROTACs.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Collaborations Pharmaceuticals, Inc, Raleigh, NC 27606, United States.
- Department of Environmental and Molecular Toxicology, Sinnhuber Aquatic Research Laboratory, Oregon State University, Corvallis, OR 97333, United States.
- Department of Biological Sciences, Toxicology Program, North Carolina State University, Raleigh, NC 27695-7633, United States.
ORCID-Profile: Lisa Truong, Robyn L Tanguy, Sean Ekins
Förderung
- NIEHS NIH HHS
- National Institute of Environmental Health Sciences (NIEHS)
- NIGMS NIH HHS
- NIH National Institute of General Medical Sciences (NIGMS)
- NIH HHS
- NIH
Zitieren
APA
Yogodzinski, C., Harris, J. S., Lane, T. R., Barnes, M., Vignaux, P. A., Raman, R., Truong, L., Tanguy, R. L., Kullman, S. W., & Ekins, S. (2026). Machine learning modeling of zebrafish toxicity endpoints after exposure to PROTACs. Toxicological Sciences, 209(1). https://doi.org/10.1093/toxsci/kfaf162
Vancouver
Yogodzinski C, Harris JS, Lane TR, Barnes M, Vignaux PA, Raman R, et al. Machine learning modeling of zebrafish toxicity endpoints after exposure to PROTACs. Toxicological Sciences. 2026;209(1). doi:10.1093/toxsci/kfaf162
BibTeX
@article{Yogodzinski2026,
author = {Yogodzinski, Christopher and Harris, Joshua S and Lane, Thomas R and Barnes, Morgan and Vignaux, Patricia A and Raman, Renuka and Truong, Lisa and Tanguy, Robyn L and Kullman, Seth W and Ekins, Sean},
title = {Machine learning modeling of zebrafish toxicity endpoints after exposure to PROTACs},
journal = {Toxicological Sciences},
year = {2026},
volume = {209},
number = {1},
doi = {10.1093/toxsci/kfaf162}
}
EUVIMED