Related Experiment Video
Updated: Jul 19, 2025
12:03
Two-Dimensional Visualization and Quantification of Labile, Inorganic Plant Nutrients and Contaminants in Soil
Published on: September 1, 2020
6.2K
Averaging Strategy for Interpretable Machine Learning on Small Datasets to Understand Element Uptake after Seed
Hengjie Yu1,2, Shiyu Tang3, Sam Fong Yau Li3
1College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China.
Environmental Science & Technology
|August 18, 2023
Summary
Plant uptake of nanoparticles is key for sustainable seed nanotreatment. Solubility, size, and surface area of nanoparticles significantly influence their concentration in maize seedlings, enabling safer nanomaterial design.
Area of Science:
- Agricultural Science
- Environmental Science
- Nanotechnology
Background:
- Understanding nanomaterial behavior in plants is vital for safe agricultural applications.
- Seed nanotreatment offers a promising method for delivering nanoparticles to crops.
Purpose of the Study:
- To predict and explain the relative metal/metalloid concentration (RMC) in maize seedlings after seed priming with nanoparticles.
- To develop interpretable machine learning models for understanding nanoparticle uptake and translocation.
Main Methods:
- Collected a dataset of 280 experimental instances on nanoparticle-primed maize seedlings.
- Applied an averaging strategy and interpretable machine learning (LightGBM, RuleFit) for prediction and explanation.
- Developed a visualization tool (RuleGrid) to illustrate feature effects and interactions.
Main Results:
- Solubility of nanoparticles strongly correlated with model prediction performance.
- Key factors influencing RMC include nanoparticle solubility, surface area, concentration, zeta potential, hydrodynamic diameter, seedling part, and plant weight.
- Self-interpretable RuleFit models accurately predicted RMC using six identified features.
Conclusions:
- The study provides an interpretable, data-driven approach to understand nanoparticle fate in plants.
- Findings contribute to the safety-by-design of nanomaterials for agricultural and environmental applications.
- Consistent relationships between parameters and RMC were confirmed across different methods.
