Few-Shot Prediction of Toxicity of Ionic Liquids Supported by Attentive Model-Agnostic Meta-Learning
Chemical Research in Toxicology · 15.06.2026 · 39 (6) · 1123-1140
DOI: 10.1021/acs.chemrestox.6c00022 ↗Kurz gefasst
Fachartikel in der Zeitschrift Chemical Research in Toxicology, online erschienen am 14.05.2026. Verfasst von Karol Baran und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Physical Chemistry, Faculty of Chemistry, Gdansk University of Technology, Narutowicza Str. 11/12, 80-233 Gdansk, Poland. Hauptthemen laut MEDLINE: Ionic Liquids, Machine Learning. Die Arbeit nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Narodowe Centrum Nauki.
Abstract (Originalsprache)
Prediction of chemical compounds’ toxicity enables efficient and rapid screening at the cost of utilizing experimental data as a foundation for artificial intelligence (AI) models. Given the constraints of limited data availability, few-shot learning techniques, like those based on meta-learning models, may be beneficial. In this study, the ionic liquid (IL) toxicity dataset was employed. Initially, models underwent pretraining on data from related domains before being adapted in a few-shot manner, adhering to the Model-Agnostic Meta-Learning (MAML) algorithm. The MAML algorithm utilized a neural network to predict toxicity based on the descriptors of the molecular structure. MAML enhanced the accuracy of IL toxicity prediction even with limited data, particularly when similar tasks were provided during meta-training, such as between Escherichia coli and Vibrio fischeri . Its effectiveness, however, is dependent on the diversity of the chemical space of the task for which adaptation is performed. As the final part of the study, a new version of MAML (i.e., attentive MAML) was proposed to address the limitations of the basic method. In our Att-MAML modified version of the algorithm, a part of the neural network was dedicated to adapting the importance of latent features with respect to the given test task. This approach allowed a significant reduction in standard deviation and overall improvement of the performance metrics in the low-data regime. This work advances the toxicity prediction in extremely low-data regimes beyond the current state of the art. Incorporation of attentive latent feature scaling (with Att-MAML) handles the limitations inherent to extremely low-data regimes. This novel method allows reductions in the standard deviation of metric values even while adapting on sets as small as four records.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Physical Chemistry, Faculty of Chemistry, Gdansk University of Technology, Narutowicza Str. 11/12, 80-233 Gdansk, Poland.
- Department of Dental Prosthetics, Medical University of Gdansk, Sklodowskiej-Curie St. 3c, 80-210 Gdansk, Poland.
ORCID-Profile: Karol Baran
Förderung
- Narodowe Centrum Nauki
Zitieren
APA
Baran, K., Derron, T., Eichenlaub, J., & Kloskowski, A. (2026). Few-Shot Prediction of Toxicity of Ionic Liquids Supported by Attentive Model-Agnostic Meta-Learning. Chemical Research in Toxicology, 39(6), 1123-1140. https://doi.org/10.1021/acs.chemrestox.6c00022
Vancouver
Baran K, Derron T, Eichenlaub J, Kloskowski A. Few-Shot Prediction of Toxicity of Ionic Liquids Supported by Attentive Model-Agnostic Meta-Learning. Chemical Research in Toxicology. 2026;39(6):1123-1140. doi:10.1021/acs.chemrestox.6c00022
BibTeX
@article{Baran2026,
author = {Baran, Karol and Derron, Tomasz and Eichenlaub, Joachim and Kloskowski, Adam},
title = {Few-Shot Prediction of Toxicity of Ionic Liquids Supported by Attentive Model-Agnostic Meta-Learning},
journal = {Chemical Research in Toxicology},
year = {2026},
volume = {39},
number = {6},
pages = {1123--1140},
doi = {10.1021/acs.chemrestox.6c00022}
}
EUVIMED