The Spectroscopy in the Middle Infrared Region (MIR) is fast and efficient alternative for analysis and quality control of dairy products. The Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) are unsupervised chemometric pattern recognition that establish sample groups based on similarities between them.
The objective was to test the efficiency of the PCA and HCA methods for identification and classification of different serum samples.
Samples were obtained from buttermilk, whey raw, pasteurized whey, whey ricotta, fresh cheese whey and whey of mozzarella cheese from the Triangulo Mineiro region, Minas Gerais, Brazil. The samples were stored under refrigeration at 5.0 ° C.
MIR Measurements were performed at intervals 48 hours for a total period of 10 days, in SpectrumTwo FT-IR spectrometer (Perkin Elmer) with accessory horizontal attenuated total reflectance (HATR) of zinc selenide (ZnSe) on 4000-600 cm-1, 4 cm-1 resolution and 16 scans per spectrum, totaling 90 spectra.
The Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) were performed using MATLAB (version 6.1, The Mathworks, Natick, MA, USA).
The dendrogram generated by HCA method created six distinct groups: of buttermilk samples, the raw whey samples, ricotta whey samples, pasteurized whey samples, of fresh cheese whey samples and whey of mozzarella samples.
The principal components PC1, PC2 and PC3 PCA, capturing 70.14%, 24.63% and 4.11% of the total variance, respectively, totaling 98.88% of variance captured. Six well-defined groups were formed.
The chemometric analysis exploratory PCA and HCA, were effective in the classification of serum samples of different origins.