Related Experiment Video
Updated: Sep 10, 2025
Rapid Homogeneous Detection of Biological Assays Using Magnetic Modulation Biosensing System
Published on: June 13, 2010
Machine Learning-Driven Dual-Recognition Magnetic Imprinted Polymers: Host-Guest/Aptamer Synergy Enabling
Jiaming Zhang1, Yanxia Ma1, Jinbo Cao1
1Department Guangzhou Key Laboratory of Analytical Chemistry for Biomedicine, GDMPA Key Laboratory for Process Control and Quality Evaluation of Chiral Pharmaceuticals, School of Chemistry, South China Normal University, Guangzhou 510006, Guangdong, China.
This study introduces a novel magnetic solid-phase microextraction coupled with HPLC system for detecting chloramphenicol in food. Machine learning optimizes the process, enhancing detection accuracy and efficiency for food safety.
Area of Science:
- Analytical Chemistry
- Nanotechnology
- Biotechnology
Background:
- Detecting chloramphenicol (CAP) in complex food matrices presents significant analytical challenges.
- Existing methods often lack sensitivity, specificity, or efficiency for real-world food samples.
Purpose of the Study:
- To develop an advanced analytical system for sensitive and accurate CAP detection in food.
- To integrate machine learning and molecular recognition for optimizing the extraction and detection process.
Main Methods:
- Fabrication of a nanocomposite material (AC-MSF) using magnetic nanoparticles, carboxylated pillar[5]arene, and aptamers for dual molecular recognition.
- Application of Bayesian optimization and six machine learning models (XGBoost, SVM) to optimize polymerization and extraction parameters.
- Utilizing SHAP analysis to identify critical parameters and kinetic simulations to elucidate the recognition mechanism.
Main Results:
- Optimized conditions using ML reduced cross-linker usage by 22.2% and polymerization time by 57%.
- Demonstrated synergistic dual recognition mechanism between CAP, pillararene, and aptamer.
- Achieved a low detection limit of 0.69 μg/L with high recoveries (85.0%-96.2%) in honey, milk, and egg samples.
Conclusions:
- The developed Apt-MIP-CP[5]A@SiO2@Fe3O4 (AC-MSF) system offers a highly efficient and sensitive method for CAP detection in food.
- Machine learning integration significantly improves analytical method optimization, reducing resource consumption.
- The synergistic dual recognition strategy provides a robust platform for detecting small molecules in complex matrices.
