Related Experiment Video
Updated: Jun 3, 2026
09:28
Engineering Molecular Recognition with Bio-mimetic Polymers on Single Walled Carbon Nanotubes
Published on: January 10, 2017
Machine Learning-assisted Raman Spectral Analysis of Serotonin-responsive ssDNA-SWCNT Nanosensor for Improved
Yunseo Choe1, Myeongkee Park2, Sanghwa Jeong3
1School of Biomedical Convergence Engineering, Pusan National University.
Journal of Visualized Experiments : Jove
|June 1, 2026
Summary
This study introduces a new method using Raman spectroscopy and machine learning to accurately detect serotonin (5-HT) in real-time. This approach overcomes previous limitations in distinguishing serotonin from dopamine, enabling more reliable neurotransmitter sensing.
Area of Science:
- Neuroscience
- Materials Science
- Analytical Chemistry
Background:
- Serotonin (5-hydroxytryptamine, 5-HT) is crucial for neuromodulation.
- Existing detection methods lack real-time sensitivity and selectivity for 5-HT.
- A previously developed near-infrared serotonin nanosensor (nIRHT) using single-walled carbon nanotubes (SWCNTs) showed sensitive 5-HT detection but could not differentiate it from dopamine (DA).
Purpose of the Study:
- To overcome the selectivity challenge in nIRHT-based 5-HT detection.
- To develop a method for discriminating between 5-HT and DA using Raman spectroscopy and machine learning.
- To enhance the practical application of SWCNT-based nanosensors for neurotransmitter sensing.
Main Methods:
- Utilized Raman spectroscopy to analyze G-band spectral features of nIRHT upon binding with 5-HT and DA.
- Employed differential Raman (ΔRaman) to isolate analyte-specific spectral changes.
- Trained and evaluated three machine learning models (including random forest) for classification of neurotransmitters.
Main Results:
- Distinct G-band spectral signatures were observed for 5-HT versus DA binding to nIRHT.
- DA binding caused a greater G-band suppression compared to 5-HT.
- The random forest model using ΔRaman achieved 95.8% accuracy, significantly outperforming models using raw spectra.
- The method demonstrated high specificity, with negligible responses to other neurotransmitters like acetylcholine, GABA, and glutamate.
- Achieved a detection limit of 0.1 µM, suitable for physiological applications.
Conclusions:
- Raman spectroscopy combined with machine learning effectively addresses the selectivity limitations of the nIRHT fluorescence sensor.
- This approach transforms the nIRHT into a platform for robust neurotransmitter discrimination.
- The developed method offers a promising solution for sensitive and selective real-time neurotransmitter sensing in physiological contexts.
