Application of artificial neural networks to develop a classification model between genetically modified maize (Bt-176) and conventional maize by applying lipid analysis data

The main objective of this study was to introduce the artificial neural network (ANN) technique into the field of food analysis. The specific purpose was to evaluate the lipid distribution of Bt-176 transgenic maize compared to that of conventional maize. The crude oil extracted from the grains of genetically modified maize (Bt-176) and nontransgenic maize was characterized in terms of the fatty acid, sterol, tocopherol distribution as well as the lipid classes and unsaponifiable level. The content of total lipids was within the range of 3.21–3.40% of grain dry matter. Fractionation of lipids into polar and nonpolar classes showed that the transgenic maize (Bt-176) contained more polar lipids than the control maize. In general, results obtained from lipid distribution analysis showed that, except for a few minor differences, the grains of Bt-176 maize were comparable in composition to that of the control maize. On the other hand, the analytical data have been elaborated by supervised pattern recognition technique (ANN) in order to classify genetically modified maize (Bt-176) and conventional maize as well as to authenticate the origin of the samples.