Rapid single grain protein analysis

Protein content is one of the most important quality factors in wheat as it influences the technological performance in baked products in combination with protein quality, which is determined by varietal choice.

Near-infrared (NIR) spectroscopy is widely used for rapid measurement of protein content. Conventional instruments measure the average value for a bulk sample, but little is known about the distribution of protein content between individual kernels.

Whole-grain applications are attractive to breeding programmes and industry due to the non-destructive scanning of samples and the speed of analysis, with the possibility of on-line or in-line data acquisition.

Our study used hyperspectral imaging (HSI) to make rapid measurements of protein content for individual wheat kernels. HSI combines NIR spectroscopy and digital imaging to give information about the chemical properties of objects and their spatial distribution. Partial least squares (PLS) regression was used to develop a calibration for protein content using reference measurements for 2000 kernels. The study assessed the typical uniformity present in commercial wheat samples and applied the calibration to visualise the protein distribution within single kernels.