LAUSR

BACKGROUND Distiller's dried grains with solubles (DDGS) are coproducts of the biofuels industries that use corn as raw material. This cereal is commonly contaminated by mycotoxins, including fumonisins (FBs), which can pose a serious health threat to humans and animals. Corn DDGS are typically used as a protein-rich animal feed. As mycotoxins from the original cereal grains become concentrated in DDGS, mycotoxicological monitoring is highly required before their use as ingredient in the industry. RESULTS This work aimed to develop a methodology for predicting fumonisins B1 (FB1 ) and B2 (FB2 ) in corn DDGS using near infrared reflectance spectroscopy (NIRS) technology associated with chemometric methods. One hundred and ninety corn DDGS samples originating from Brazilian ethanol plants and feed mills were included in this assessment. Two datasets were created, one for calibration (132 samples) and another for external validation (58 samples). Partial least squares regression and a cross-validation approach were applied to build the models. Liquid chromatography coupled to tandem mass spectrometry was used as the reference methodology. Calibration results of correlation coefficient and residual prediction deviation for FB1 and FB2 were, respectively: 0.90 and 0.88; and 2.16 and 2.06. CONCLUSIONS Values of the external validation dataset were compared and no statistical difference was found between groups, thus indicating a satisfactory predictive ability and confirming the potential of NIRS to predict fumonisins in corn DDGS. This article is protected by copyright. All rights reserved.