Geographic Origin Traceability and Stable Isotope Prediction of Pgi Hongyuan Yak Milk by Ftir

Abstract

The geographical origin of Protected Geographical Indication (PGI) Hongyuan yak milk and three other non-PGI milk samples were distinguished by Fourier Transform Infrared Spectroscopy (FTIR) combined with chemometrics analysis, which was also used to predict δ13C, δ15N, δ2H, and δ18O values in the milk samples. Principal component analysis (PCA), orthogonal partial least squares discriminant analysis (OPLS-DA), linear discriminant analysis (LDA), and support vector machine regression (SVMR) were used for data analysis. The classification accuracies of LDA and OPLS-DA reached 90.9% and 94.1%, respectively. Meanwhile, the best SVMR calibration models for predicting the four stable isotopes of milk (δ13C, δ15N, δ2H, and δ18O) were moderately to well predicted by FTIR (R2: 0.62 to 0.75). The results show that FTIR combined with chemometrics provides a rapid and non-destructive method for milk origin differentiation and δ13C, δ15N, δ2H, and δ18O prediction.

Keywords: Food authentication, Protected geographical indications, Milk, FTIR, Stable isotope