Multi species-toxicity prediction for metallic nanomaterials
2023
Beihang University, Beijing, China
The wide production and use of metallic nanomaterials (MNMs) leads to increased emissions into the aquatic environments and induces high potential risks. The objective of this study was to develop a machine learning-based regression model for aquatic toxicity prediction that takes into account physicochemical properties of MNMs, environmental factors, and different organisms with their own traits and exposure conditions. To achieve this, a model of 14 different MNMs against 51 species was developed based on published data sets, and the model was validated against data obtained from recently published literature. The model was used to analyse the importance and interaction between physicochemical properties, environmental factors and species. Feature importance and interaction analysis indicated that exposure duration, illumination, primary size, and hydrodynamic diameter were the main factors affecting the ecotoxicity of MNMs.
The in-silico approach enables the multi species-toxicity prediction for MNMs and will help to further explore exposure pathways.
Using machine learning to predict adverse effects of metallic nanomaterials to various aquatic organisms
Ying Wang, Wenhong Fan
Added on: 02-06-2024
[1] https://pubs.acs.org/doi/10.1021/acs.est.2c07039?ref=PDF
