Related Experiment Video
Updated: Jun 5, 2025
10:20
Quantification of Heavy Metals and Other Inorganic Contaminants on the Productivity of Microalgae
Published on: July 10, 2015
15.9K
Machine learning-based prediction and model interpretability analysis for algal growth affected by microplastics
Wenhao Li1, Xu Zhao1, Xudong Xu1
1Key Laboratory of Pollution Processes and Environmental Criteria (Ministry of Education), Tianjin Key Laboratory of Environmental Remediation and Pollution Control, College of Environmental Science and Engineering, Nankai University, Tianjin 300350, China.
The Science of the Total Environment
|December 15, 2024
Summary
Machine learning models predict microplastic (MP) toxicity on microalgae. Exposure time, MP concentration, and size significantly impact algal growth, with smaller MPs and longer exposure times showing greater effects.
Area of Science:
- Environmental Science
- Ecotoxicology
- Computational Biology
Background:
- Microplastics (MPs) are pervasive pollutants in aquatic environments.
- MPs pose significant toxicity risks to aquatic organisms, particularly microalgae.
Purpose of the Study:
- To develop machine learning models for predicting MP effects on algal growth.
- To identify key features influencing MP toxicity using model interpretability.
Main Methods:
- Literature data compilation of 408 samples.
- Implementation of Random Forest, CatBoost, and LightGBM algorithms.
- Construction of classification models for algal growth prediction.
Main Results:
- LightGBM model achieved the highest accuracy (0.8305) and Kappa (0.7165).
- Key influential features identified: exposure time, MP concentration, and MP size.
- Algal growth inhibition increased with exposure duration (72-216 h), MP concentration (0-300 mg/L), and smaller MP size relative to algal cells.
Conclusions:
- Successfully developed predictive models for MP impact on algal growth.
- Provided insights into MP toxicity mechanisms and influencing factors.
- Highlighted the importance of exposure parameters and MP characteristics in ecotoxicological assessments.
