Abstract
Tryptophan fluorescence spectra of eight different soft cheeses were recorded directly on cheese samples using front-face fluorescence spectroscopy. Discriminant ability of the data was investigated by discriminant analysis. A correct classification was observed for 95% and 92% of the calibration and validation samples, respectively. It was concluded that tryptophan fluorescence spectra enable the identity of individual cheeses to be finger-printed. Canonical correlation analysis was applied to soft-cheese sensory profile data and fluorescence spectral collection in order to measure the link between the two groups of variables measured on the same samples. The two groups of variables were found highly correlated since the squared canonical coefficients for canonical variates 1 and 2 were 0.93 and 0.80, respectively. A subset of four cheeses was investigated closely in order to establish a molecular basis of the discrimination. It was shown that molecular level information may be derived from the fluorescence spectra.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have