Moisture effects on robustness of sorghum grain protein near‐infrared spectroscopy calibration

Abstract

Background and objectives A near‐infrared (NIR) spectroscopy method was developed for rapid and nondestructive evaluation of protein content of intact sorghum grains. Effect of grain sample moisture variation on the robustness of protein calibration was investigated. Findings An initial NIR protein calibration model with sorghum grains in 7.28%–11.57% moisture content range with coefficient of determination (R2) of 0.94 and standard error of cross‐validation (SECV) of 0.41%, predicted protein content of an external validation set of different varieties with R2 = 0.90, root‐mean‐square error of prediction (RMSEP) = 0.42% and bias = −0.02%. However, when grains with a wider range of moisture content (7.50%–17.75%) were used for validation, prediction errors increased with R2 = 0.72 RMSEP = 0.87% and bias = −0.32%. Inclusion of grains with a wider moisture range to the calibration set improved the performance of the calibration model with a R2 = 0.83, RMSEP = 0.67% with a bias of −0.04%. Conclusions Variation of moisture content in grains affected the performance of the NIR protein calibration model. Likewise, inclusion of moisture variation in the calibration sample set improved the robustness of the model. Significance and novelty In addition to the traits of interest, variation of other physical or chemical traits should also be considered for inclusion into the calibration sample set to improve the robustness of NIR calibration models. This is especially important when NIR spectroscopy methods are developed for evaluation of breeding populations as the future grain samples from numerous crosses may be substantially diverse.

Extended multiplicative signal correction to improve prediction accuracy of protein content in weathered sorghum grain samples

“…Peiris et al (2019) showed that moisture content of sorghum grains affected the performance of the NIR protein calibration model and inclusion of moisture variation in the calibration sample set improved the robustness of the model. Therefore, the C1 samples intended for the development of a robust NIR calibration for protein content determination in sound sorghum grain samples had a moisture range from 7.28% to 18.68% (Table 2).…”

“…This may hold true when developing NIR calibrations for other sorghum traits as well and thus warrants further research in developing calibrations for assessing other traits of weathered sorghum. The most important NIR bands carrying the information on sorghum protein is found in the 950–1200 nm region of the spectra (Peiris et al., 2019). Since weathered grains show spectral distortions in this spectral region, correction of such distortions in spectra by EMSC preprocessing improved the prediction accuracy of the NIR protein calibration when applied to weathered grains allowing for NIRS to be used on sample sets containing weathered sorghum grain samples.…”

“…NIR methods have been developed for several components of sorghum grain including protein content (Peiris, Bean, Chilual, Perumal & Jagadish, 2019). In addition to the variation in protein content, variations in other chemical and physical properties of the grain samples influence the robustness of the NIR method.…”

“…Since water is a strong NIR absorber (Büning‐Pfaue, 2003), grain moisture levels significantly influence the spectral features. Addition of grain moisture variability to the calibration sample set improved the robustness of NIR calibration for sorghum protein content (Peiris et al, 2019). As weathering of sorghum grains can alter chemical and physical properties of grain samples, and more importantly properties of the surface of the grains, NIR spectra of such grain samples may be distorted compared to spectra of sound grain samples.…”

Potential food applications of sorghum (Sorghum bicolor) and rapid screening methods of nutritional traits by spectroscopic platforms

“…The crude protein content and specific amino acids have been analyzed by NIR (Peiris et al, 2019). The crude protein content is comparatively higher in sorghum than wheat, corn, and barley, and morphologically, sorghum grains are not covered by a protective layer.…”