Prediction of selected rheological characteristics of wheat based on Glutopeak test parameters

Abstract

The GlutoPeak®-Test, a new rapid small-scale technique, was proposed as an alternative method for evaluation of wheat grain and tool for predicting the wheat flour quality. Samples obtained from an industrial mill were analyzed by a GlutoPeak test (whole grain flours) as well as by farinograph and extensograph tests (refined flours). Firstly, linear correlation coefficients between water absorption, dough stability, dough energy and defined parameters of GlutoPeak were calculated. Next, a sequential multiple quadratic regression analysis (backward, forward and stepwise), a logistic regression analysis and a PLSR analysis were applied. The correlation between the flour water absorption and most of the parameters obtained from the GlutoPeak test were strong (r ≥ 0.74, p < 0.001). For stability, the r value was 0.40, while for energy it was 0.44. Based on the obtained results it could be state that in the case of water absorption, the best fit was the sequential regression model, for dough stability sequential regression model and the PLSR model had the best fit, whereas logistic regression model was the best fitted to the energy. Unfortunately, after cross validation it was found that the last model is not good enough for energy prediction.