Rapid Inference of Distinct Genotypes in Vibrio parahaemolyticus by SVM-Assisted Surface-Enhanced Raman Spectroscopy

Vibrio parahaemolyticus is a kind of Gram-negative bacteria widely distributed in estuarine and coastal areas. It was discovered in dried mackerel in Osaka and has now become one of the main pathogens causing foodborne diseases. The seafood contamination caused by it poses a serious threat to human health and safety. Although most V. parahaemolyticus strains are non-pathogenic, they are often ignored in the routine detection process. The thermostable direct hemolysin (TDH) and TDH-related hemolysin (TRH) encoded by the tdh and trh genes, respectively, are considered its main virulence factors. Different genotypes of strains are closely related to antibiotic resistance phenotypes. Therefore, the related tdh and trh genes are considered the main virulence markers in the epidemiology of V. parahaemolyticus . Here, based on label-free surface-enhanced Raman spectroscopy (SERS), V. parahaemolyticus with different tdh and trh expression were directly detected. By combining support vector machine (SVM), the test samples were divided into four categories according to different genotypes, and the inference of different genotypes from phenotypic spectra was achieved within 7 min, and the classification accuracy of the model was 98.46%. The results of this work demonstrate that the combination of label-free SERS spectroscopy and SVM algorithms has great potential in the field of bacterial classification.