Related Experiment Video
Updated: Jul 17, 2025
Author Spotlight: Advancing SERS Technology: Au@Carbon Dot Nanoprobes for Label-Free Analysis and Imaging
Published on: June 9, 2023
Rapid classification of SARS-CoV-2 variant strains using machine learning-based label-free SERS strategy
Jingwang Qin1, Xiangdong Tian1, Siying Liu2
1State Key Laboratory of Structural Chemistry, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian, 350002, PR China; Department of Translational Medicine, Xiamen Institute of Rare Earth Materials, Haixi Institute, Chinese Academy of Sciences, Xiamen, 361021, PR China.
Rapidly identifying severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants is key to controlling the pandemic. This study uses surface-enhanced Raman scattering (SERS) and machine learning (ML) for quick and accurate variant detection.
Area of Science:
- Biotechnology
- Virology
- Data Science
Background:
- Continuous mutation of SARS-CoV-2, particularly in the spike (S) protein, drives viral transmission and immune evasion.
- Accurate and rapid identification of SARS-CoV-2 variants is critical for effective public health interventions and treatment strategies.
Purpose of the Study:
- To develop and validate a novel method for the rapid and accurate identification of SARS-CoV-2 variants.
- To assess the efficacy of label-free surface-enhanced Raman scattering (SERS) combined with machine learning (ML) for variant discrimination.
Main Methods:
- Establishment of a SERS spectral database for various SARS-CoV-2 variants.
- Development of a logistic regression (LR) machine learning model for variant classification.
- Testing the model's accuracy on known variants and blind tests of human nasal swabs.
Main Results:
- The developed classifier achieved 100% accuracy in identifying Beta, Delta, Wuhan, and Omicron (BA.1) variants.
- The method demonstrated 100% accuracy in blind tests on positive and negative human nasal swabs.
- The SERS-ML approach provided accurate variant identification within 10 minutes.
Conclusions:
- Machine learning-based SERS technology offers a promising solution for the accurate and rapid discrimination of SARS-CoV-2 variants.
- This technology has the potential for use in real-time diagnosis and guiding therapeutic decisions.
- The method is effective in complex biological samples, facilitating clinical applications.
