On June 15, 2024, Professor Liu Jing from Shanghai University of Technology and a team led by Jin Chengni and Yan Shuaishuai from Shanxi Normal University published a paper titled "Raman Activated Cell Ejection for Validating the Reliability of the Raman Fingerprint Database of Foodborne Pathogens" in the journal Foods. The study predicted the types of foodborne pathogens using Raman detection combined with machine learning algorithms, and validated the accuracy of the method through single-cell sorting combined with 16S identification and metagenomic sequencing technology. Changguang Chenying's core products, P300 laser confocal Raman spectrometer and PRECI SCS microbial single-cell sorter, have provided powerful tools for the detection and sorting validation of foodborne pathogens in this study.

Abstract

Raman spectroscopy for rapid identification of foodborne pathogens based on phenotype has attracted increasing attention, and the reliability of the Raman fingerprint database through genotypic determination is crucial. In the research, the classification model of four foodborne pathogens was established based on t-distributed stochastic neighbor embedding (t-SNE) and support vector machine (SVM); the recognition accuracy was 97.04%. The target bacteria named by the model were ejected through Raman-activated cell ejection (RACE), and then single-cell genomic DNA was amplified for species analysis. The accuracy of correct matches between the predicted phenotype and the actual genotype of the target cells was at least 83.3%. Furthermore, all anticipant sequencing results brought into correspondence with the species were predicted through the model. In sum, the Raman fingerprint database based on Raman spectroscopy combined with machine learning was reliable and promising in the field of rapid detection of foodborne pathogens.

The paper links：

https://www.mdpi.com/2304-8158/13/12/1886