A novel approach to discriminate transgenic from non-transgenic soybean oil using FT-MIR and chemometrics

摘要

A methodology was developed to distinguish transgenic from non-transgenic soybean oils samples by using FT-MIR spectroscopy coupled with discrimination techniques, including Soft Independent Modeling of Class Analogies (SIMCA), Support Vector Machine-Discriminant Analysis (SVM-DA) and Partial Least Squares, Discriminant Analysis (PLS-DA). The discrimination success rate of these three methods was compared, and different types of preprocessing were investigated. Based on the results, the best option was PLS-DA with a 100% rate of discrimination, independent of the preprocessing method used. (C) 2014 Elsevier Ltd. All rights reserved.