Skip to main content
ARS Home » Plains Area » Manhattan, Kansas » Center for Grain and Animal Health Research » Stored Product Insect and Engineering Research » Research » Publications at this Location » Publication #433155
Research Project: Systems-Based Approaches for Mitigating Losses from Stored Product Insects

Location: Stored Product Insect and Engineering Research

Title: Near-infrared spectroscopy for the single-kernel analysis of sorghum protein content

Author
item MENDOZA, PRINCESSTIFFANY - Kansas State University
item ARMSTRONG, PAUL - Retired ARS Employee
item Scully, Erin
item Wu, Xiaorong
item PEIRIS, KAMARANGA - Retired ARS Employee
item BEAN, SCOTT - Retired ARS Employee
item SILIVERU, KALIRAMESH - Kansas State University

Submitted to: Sensors
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/4/2026
Publication Date: 5/7/2026
Citation: Mendoza, P.D., Armstrong, P.R., Scully, E.D., Wu, X., Peiris, K.H., Bean, S.R., Siliveru, K. 2026. Near-infrared spectroscopy for the single-kernel analysis of sorghum protein content. Sensors. https://doi.org/10.3390/s26102936.
DOI: https://doi.org/10.3390/s26102936

Interpretive Summary: Protein content is a vital quality trait in sorghum that influences its utility in human and animal food products, market value, and other end-uses. Different sorghum cultivars have a wide variation in protein content, which can also be influenced by the environment and by genetics. Moreover, protein content can vary extensively even within grains collected from the same plant. To rapidly quantify protein content in single kernels for breeding purposes, this study developed a method for a non-destructive and rapid prediction of protein content in individual sorghum kernels using single-kernel near-infrared spectroscopy (SKNIR) couple with machine learning-based data analysis to identify spectra associated with protein content. The highest performing model resulted in a standard error of only 0.83%, meaning that predicted protein levels were very similar to actual values. The relative predictive determinant was 3.40, indicating that the model performs well on unknown samples. Altogether, these results show that the model performs well and could be applied to accelerate the screening of sorghum varieties by protein content for breeding and quality applications.

Technical Abstract: Protein content is a vital quality trait in sorghum that influences breeding approaches, end-use applications, and market value. Influenced by genetic, agronomic, and environmental variability, sorghum is characterized by its wide variation in composition, which may also be evident in kernels from the same sample. This study developed and evaluated a method for a non-destructive and rapid prediction of protein content in individual sorghum kernels using single-kernel near-infrared spectroscopy (SKNIR). Applying different pre-processing techniques to the spectra collected from intact kernels, the calibration models were developed using partial least squares regression and the reference protein content values obtained from the LECO combustion method. The best model was obtained using multiplicative scatter correction as pre-processing, resulting in a standard error of prediction of 0.83% and a relative predictive determinant of 3.40. These were indicative of the good predictive ability of the model and the instrument to be applied in quality control and sorting applications. These results highlight the potential of SKNIR to capture the inter-kernel variability in sorghum protein content and enhance screening for grain quality in breeding and grain processing.