Abstract
Fourier-transform infrared equipped with attenuated total reflection (ATR–FT-IR) was used in combination with multivariate statistical analysis for classification and identification of food pathogens Staphylococcus and Salmonella. The goals of the present study were to validate the feasibility of ATR–FT-IR in collecting information for discriminating different bacteria, and to assess the merits of two routes for effectively identify target foodborne bacteria. The results showed that ATR–FT-IR was able to provide enough chemical information of each species. Cluster-analysis-test was able to identify target bacteria at the genus and species level using Pearson's product-moment correction coefficient and Ward's algorithm. Partial least squares regression discriminant analysis (PLS-DA) coupled with multiplicative scatter correction (MSC), standard normal variate (SNV) and their derivatives demonstrated the probable use of this routine method to differentiate food pathogens at the sub-species level.