Skip to main content
ARS Home » Plains Area » Manhattan, Kansas » Center for Grain and Animal Health Research » Grain Quality and Structure Research » Research » Publications at this Location » Publication #360004
Research Project: Impact of the Environment on Sorghum Grain Composition and Quality Traits

Location: Grain Quality and Structure Research

Title: Moisture effects on robustness of sorghum grain protein NIR spectroscopy calibration

Author
item PEIRIS, KAMARANGA H.S. - Kansas State University
item Bean, Scott
item CHILUWAL, ANUJ - Kansas State University
item PERUMAL, RAMASAMY - Kansas State University
item JAGADISH, KRISHNA - Kansas State University

Submitted to: Cereal Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/29/2019
Publication Date: 7/8/2019
Citation: Peiris, K., Bean, S.R., Chiluwal, A., Perumal, R., Jagadish, K. 2019. Moisture effects on robustness of sorghum grain protein NIR spectroscopy calibration. Cereal Chemistry. 96: 678-688. https://doi.org/10.1002/cche.10164.
DOI: https://doi.org/10.1002/cche.10164

Interpretive Summary: A NIR spectroscopy method was developed for rapid and non-destructive evaluation of protein content of intact sorghum grains. Effect of grain sample moisture variation on the robustness of protein calibration was investigated. Variation of moisture content in grain samples 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. 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 samples from numerous crosses may be substantially diverse.

Technical Abstract: A NIR spectroscopy method was developed for rapid and non-destructive evaluation of protein content of intact sorghum grains. Effect of grain sample moisture variation on the robustness of protein calibration was investigated. An initial NIR protein calibration model with coefficient of determination (R2) of 0.95 and standard error of cross validation (SECV) of 0.36%, predicted protein content of an external validation set of different varieties with R2 = 0.89, Root Mean Square Error of Prediction (RMSEP) = 0.43 % and bias = 0.02 %. However, when samples with a wider range of moisture content (7.68 to 18.68 %) was used for validation, prediction errors increased with the calibration having R2 = 0.44, RMSEP = 1.54 % and bias = 0.95 %. Inclusion of grain samples with a wider moisture range to the calibration set improved the performance of the calibration model with a R2 = 0.83, RMSEP = 0.64 % with a bias of -0.15 %. Variation of moisture content in grain samples 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. 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 samples from numerous crosses may be substantially diverse.