Discrimination of almonds (amygdalus) with respect to their genotype by using Fourier Transform Infrared Spectroscopy and chemometrics

Fourier Transform Infrared Spectroscopy (FT-IR) is a well-established analytical tool in the chemical industry and analytical laboratory. Its utility arises from the generally well-resolved absorption bands found in FT-IR spectra and the consequent relative ease of chemical identification and quantitation. However, the application of this technique to food samples has only been seriously addressed recently due to developments in instrument design [1]. Spectroscopic techniques as generally applied to authenticity issues are non-selective, i.e. they do not detect the presence or absence of a single marker compound. Rather spectra contain information about the complete chemical composition and physical state of the material under analysis [2]. In this study, Fourier Transform Infrared spectroscopy (FT-IR) coupled to chemometrics is used to develop a fast and simple method for discriminating sweet, peanut, Prunus, Pistacia atlantica and bitter almonds (amygdalus). Spectra were recorded in the range of 400–3500 cm-1, and taking 15 scans per sample. The absorbance was computed against a background spectrum of Spectralon. The reflection window plate was carefully cleaned with a soft tissue to eliminate the presence of residues between measurements. FT-IR spectroscopy and an unsupervised pattern-recognition method, Principal component analysis (PCA), Linear discriminant analysis (LDA) based on the PCA and supervised counter propagation artificial neural network (CP-ANN) groupings, models were built to discriminate each types of almonds, show excellent discrimination between the almond groups and obtaining high levels of sensitivity and specificity for each classes, with more than 97% of the samples correctly classified and discriminated.