Identification of botanical and geographical origin of distillers dried grains with solubles by near infrared microscopy

Nowadays there is a strong need for methods capable of differentiating distillers dried grains with solubles (DDGS) according to their geographical and/or botanical origin. In this study, near infrared microscopy (NIRM) has been applied to differentiate corn and wheat DDGS samples and to differentiate corn DDGS from USA, Europe and China. Two multivariate prediction models have been developed for the discriminations. The models have been validated by cross validation and with an external set of samples and have been used to predict the botanical and geographical origin. Furthermore the probability of wrong prediction of the models has been calculated, to assess their performance profile. The prediction showed very good capability of the model to classify samples according to their botanical origin, whereas satisfying results were exclusively obtained when separating samples from China against the pooled set of samples from USA and Europe.