A Multitask Active Learning Framework with Probabilistic Modeling for Multi-Species Acute Toxicity Prediction
Molecules · 30.03.2026 · 31 (7) · 1144
DOI: 10.3390/molecules31071144 ↗Kurz gefasst
Fachartikel in der Zeitschrift Molecules, online erschienen am 30.03.2026. Verfasst von Tianyu Han und 7 weiteren Autor:innen. Beteiligte Einrichtung: Shanghai Key Laboratory of Power Station Automation Technology, School of Mechatronics Engineering and Automation, Shanghai University, Shanghai 2004…. Hauptthemen laut MEDLINE: Models, Statistical, Toxicity Tests, Acute, Machine Learning. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Shanghai, Key Technologies Research and Development Program, National Natural Science Foundation of China.
Abstract (Originalsprache)
Predicting acute toxicity across species is essential for early-stage drug safety evaluation. While recent efforts have primarily focused on improving predictive accuracy, they often fail to address two critical issues: the substantial divergence in toxicity mechanisms among different species, and the inherent noise present in experimental data. To bridge this gap, we introduce a Probabilistic Multitask Active Learning (PMAL) framework for multi-species acute toxicity prediction. Our framework integrates two key modules: a Probabilistic Multitask Learning (PML) component which jointly models the predictive distributions of multiple toxicity endpoints from a probabilistic viewpoint, and an Uncertainty-based Active Learning (UAL) component which strategically selects the most informative compounds for experimental annotation based on predictive uncertainty. Empirical evaluations demonstrate that PMAL surpasses state-of-the-art methods and is capable of providing well-calibrated uncertainty estimates for small molecules across diverse toxicity endpoints. Beyond advancing multi-species toxicity prediction, the core design principles of PMAL offer a generalizable paradigm for learning in noisy multi-task environments.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Shanghai Key Laboratory of Power Station Automation Technology, School of Mechatronics Engineering and Automation, Shanghai University, Shanghai 200444, China.
- School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China.
- School of Medicine, Shanghai University, Shanghai 200444, China.
- Academy of Military Medical Sciences, Beijing 100850, China.
ORCID-Profile: Tianyu Han, Jingjing Wang, Yanpeng Zhao, Ying Lin, Lu Yu, Song He, Peng Zan, Xiaochen Bo
Förderung
- Natural Science Foundation of Shanghai
- Key Technologies Research and Development Program
- National Natural Science Foundation of China
- National Natural Science Foundation of China https://ror.org/01h0zpd94
Zitieren
APA
Han, T., Wang, J., Zhao, Y., Lin, Y., Yu, L., He, S., Zan, P., & Bo, X. (2026). A Multitask Active Learning Framework with Probabilistic Modeling for Multi-Species Acute Toxicity Prediction. Molecules, 31(7), 1144. https://doi.org/10.3390/molecules31071144
Vancouver
Han T, Wang J, Zhao Y, Lin Y, Yu L, He S, et al. A Multitask Active Learning Framework with Probabilistic Modeling for Multi-Species Acute Toxicity Prediction. Molecules. 2026;31(7):1144. doi:10.3390/molecules31071144
BibTeX
@article{Han2026,
author = {Han, Tianyu and Wang, Jingjing and Zhao, Yanpeng and Lin, Ying and Yu, Lu and He, Song and Zan, Peng and Bo, Xiaochen},
title = {A Multitask Active Learning Framework with Probabilistic Modeling for Multi-Species Acute Toxicity Prediction},
journal = {Molecules},
year = {2026},
volume = {31},
number = {7},
pages = {1144},
doi = {10.3390/molecules31071144}
}
EUVIMED