Related Experiment Video
Updated: Jul 26, 2026

An Integrated Raman Spectroscopy and Mass Spectrometry Platform to Study Single-Cell Drug Uptake, Metabolism, and Effects
Published on: January 9, 2020
Simultaneous quantitative analysis of multiple metabolites using label-free surface-enhanced Raman spectroscopy and
Xianli Tian1, Peng Wang2, Guoqiang Fang3
1School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei 230026, China; Jiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, Jiangsu 215163, China.
This study presents a novel method combining Surface-Enhanced Raman Spectroscopy (SERS) and deep learning for simultaneous metabolite detection. The technique accurately quantifies multiple metabolites, offering a sensitive tool for clinical diagnostics.
Area of Science:
- Analytical Chemistry
- Biomarker Discovery
- Computational Biology
Background:
- Metabolites are crucial biomarkers for physiological and pathological states, aiding disease progression monitoring and early detection.
- Existing analytical techniques for metabolite detection can be limited in sensitivity, cost-effectiveness, or speed for simultaneous analysis.
Purpose of the Study:
- To develop and validate an advanced analytical technique for simultaneous, quantitative detection of multiple metabolites.
- To integrate label-free Surface-Enhanced Raman Spectroscopy (SERS) with deep learning and SHAP for enhanced interpretability.
- To establish a sensitive, cost-effective, and rapid method for metabolite analysis applicable to clinical diagnostics.
Main Methods:
- Fabrication of monolayer silver nanoparticle SERS substrates using a triple-phase interfacial self-assembly method.
- Development of a custom deep neural network model with multi-channel feature extraction for spectral data analysis.
- Application of SHAP (SHapley Additive exPlanations) for visual interpretative analysis of the deep learning model's predictions.
Main Results:
- Simultaneous detection and quantitative analysis of uric acid, xanthine, hypoxanthine, and creatinine with high accuracy (R² values ranging from 0.940 to 0.977).
- Demonstrated scalability of the method with consistent performance as the number of simultaneous targets increased.
- Successful capture of complex spectral information from target metabolites in mixed solutions using the fabricated SERS substrates.
Conclusions:
- The integrated SERS and deep learning approach provides a sensitive, cost-effective, and rapid platform for multi-metabolite analysis.
- The method shows significant potential for advancing clinical diagnostics and enabling personalized medicine through precise metabolite profiling.
- SHAP analysis offers valuable insights into the predictive rationale of the deep learning model, enhancing trust and understanding.
Related Concept Videos
MALDI-TOF Mass Spectrometry
Mass Spectrometry: Complex Analysis
GC–MS is a powerful hyphenated method commonly used in forensics and environmental...
Tandem Mass Spectrometry
High-Performance Liquid Chromatography: Types of Detectors

