A Curve-Fitting Approach to the near Infrared Reflectance Measurement of Wheat Flour Proteins Which Influence Dough Quality

Abstract

A curve-fitting approach to the problem of measuring the gliadin and glutenin content of flour is presented. Theoretical spectra for 100% pure gliadin, glutenin and starch are derived based on the analysis of partially purified materials and are shown to have unique characteristics. Spectral differences at 2306 nm and 2350 nm were attributed to the presence of lipid in the glutenin component and a least squares curve-fitting procedure over the range 2050–2220 nm has been used to calculate the proportion of each component present in the spectrum of a flour. High r2 values were achieved when the results were compared with the actual values as measured by size exclusion high pressure liquid chromatography. The results for gliadin ( r2 = 0.73) and glutenin ( r2 = 0.76) suggest that the curve-fitting methodology offers a simple alternative to partial least squares (PLS) and multiple linear regression (MLR) techniques for ranking samples by gliadin and glutenin content. Although there was a large bias between the actual and uncorrected predicted values for gliadin and glutenin content, the results provided a clear justification for assuming that robust calibrations for gliadin and glutenin content can be developed using both curve-fitting and PLS calibration techniques. The least squares methodology can be used as a simple technique for estimating the gliadin and glutenin content of flours where the development and maintenance of a full calibration is either impractical or not desirable.