Deep autoencoder-based behavioral pattern recognition outperforms standard statistical methods in high-dimensional zebrafish studies
PLOS Computational Biology · 2024
DOI: 10.1371/journal.pcbi.1012423 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, 2024. Verfasst von Adrian J Green und 6 weiteren Autor:innen. Beteiligte Einrichtung: Bioinformatics Research Center, Department of Biological Sciences, NC State University, Raleigh, North Carolina, United States of America. Hauptthemen laut MEDLINE: Zebrafish, Behavior, Animal. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 83 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC0). Gefördert von: NIEHS NIH HHS, NCI NIH HHS, Intramural NIH HHS.
Abstract (Originalsprache)
Zebrafish have become an essential model organism in screening for developmental neurotoxic chemicals and their molecular targets. The success of zebrafish as a screening model is partially due to their physical characteristics including their relatively simple nervous system, rapid development, experimental tractability, and genetic diversity combined with technical advantages that allow for the generation of large amounts of high-dimensional behavioral data. These data are complex and require advanced machine learning and statistical techniques to comprehensively analyze and capture spatiotemporal responses. To accomplish this goal, we have trained semi-supervised deep autoencoders using behavior data from unexposed larval zebrafish to extract quintessential “normal” behavior. Following training, our network was evaluated using data from larvae shown to have significant changes in behavior (using a traditional statistical framework) following exposure to toxicants that include nanomaterials, aromatics, per- and polyfluoroalkyl substances (PFAS), and other environmental contaminants. Further, our model identified new chemicals (Perfluoro-n-octadecanoic acid, 8-Chloroperfluorooctylphosphonic acid, and Nonafluoropentanamide) as capable of inducing abnormal behavior at multiple chemical-concentrations pairs not captured using distance moved alone. Leveraging this deep learning model will allow for better characterization of the different exposure-induced behavioral phenotypes, facilitate improved genetic and neurobehavioral analysis in mechanistic determination studies and provide a robust framework for analyzing complex behaviors found in higher-order model systems.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Bioinformatics Research Center, Department of Biological Sciences, NC State University, Raleigh, North Carolina, United States of America.
- Sciome LLC, Research Triangle Park, North Carolina, United States of America.
- Department of Environmental and Molecular Toxicology, Oregon State University, Corvallis, Oregon, United States of America.
- Predictive Toxicology Branch, Division of Translational Toxicology, National Institute of Environmental Health Sciences, Durham, North Carolina, United States of America.
ORCID-Profile: Adrian J Green, David M Reif
Förderung
- NIEHS NIH HHS
- NCI NIH HHS
- Intramural NIH HHS
- National Institute of Environmental Health Sciences
- National Cancer Institute
- the Intramural Research Program of the NIH
Zitieren
APA
Green, A. J., Truong, L., Thunga, P., Leong, C., Hancock, M., Tanguay, R. L., & Reif, D. M. (2024). Deep autoencoder-based behavioral pattern recognition outperforms standard statistical methods in high-dimensional zebrafish studies. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1012423
Vancouver
Green AJ, Truong L, Thunga P, Leong C, Hancock M, Tanguay RL, et al. Deep autoencoder-based behavioral pattern recognition outperforms standard statistical methods in high-dimensional zebrafish studies. PLOS Computational Biology. 2024. doi:10.1371/journal.pcbi.1012423
BibTeX
@article{Green2024,
author = {Green, Adrian J and Truong, Lisa and Thunga, Preethi and Leong, Connor and Hancock, Melody and Tanguay, Robyn L and Reif, David M},
title = {Deep autoencoder-based behavioral pattern recognition outperforms standard statistical methods in high-dimensional zebrafish studies},
journal = {PLOS Computational Biology},
year = {2024},
doi = {10.1371/journal.pcbi.1012423}
}
EUVIMED