Research Project: New Sensors and Methods for Phenotypic Analysis of Small Grains

Location: Food Quality Laboratory

2021 Annual Report

Objectives
Objective 1: Enable new or refine commercial viscometry, spectroscopic imaging, and physical technologies that integrate indicators of wheat endosperm integrity. • Sub-objective 1.A. Develop a physicochemical procedure that distinguishes between pre-harvest sprouting and late maturity amylase. • Sub-objective 1.B. Develop a standard reference material for the falling number procedure, to be used in official inspection operations. Objective 2: Enable new, real-time, rapid optical methods to detect and measure sprouting, mold, and black point in harvested grain. • Sub-objective 2.A. Develop spectral imaging procedures for identification of wheat seeds damaged by black point, insect, and arrested development (immaturity). • Sub-objective 2.B. Develop imaging procedures for assessing dormancy in wheat lines when challenged with conditions favoring germination.

Approach
Wheat samples from annual breeders evaluation lines will be measured for falling number (FN). Samples with FN less than 300 s will be selected for further analysis. For these samples, a minimum of 30 seeds will be bisected transversely along the seed axis, and brush-end halves will be accumulated, as will the germ-end seeds. Amylase activity will be measured for each group. Because the late maturity amylase condition produces elevated activity more uniformly throughout a seed’s aleurone layer compared to the pre-harvest sprout condition (with activity concentrated near the germ), samples with equivalent amylase activities between the halves will be identified as LMA, and those with imbalanced amylase activities will be identified as PHS. Proteolytic enzyme activity will be measured using an azocasein substrate. The results of this assay will be confirmatory on whether PHS or LMA is the cause for elevated amylase activity as the latter condition should not produce elevated proteolytic activity. As with FN, the basis of the analytical procedure to be developed will be viscometry. Four native (unmodified) starches, wheat, corn, potato and rice, will be obtained from a laboratory chemical supplier. Rather than adding enzyme to a full fixed mass to bring FN down to a practical range, pure starch masses without added amylase will be adjusted to produce a target FN of 300 s at constant volume of added water (25 mL). Preliminary studies suggest masses of 6.25 g for potato, 5.0 g for wheat and rice, and 4.0 g for corn starches. Weekly runs will be collected on all four starches (5 twin-tube runs for each) on each of two FN instruments over a three-month period. FN stability across time will be examined by calculating the among-weeks and instrument x weeks variance components for each native starch. A repeated measures ANOVA will be performed to reveal effects of starch type, time, instrument, and their interactions. Starches will also be characterized experimentally for amylase activity, amylose/amylopectin ratio, and the presumably nil contents of nitrogen and ash. An imaging system will be developed that can be used as an inspector’s assistant to grade wheat and identify the various defect categories (e.g., mold, black point, frost, heat). Image analysis will be a multistep process. A first region of analysis (ROI) will encompass the section of the grain that is exclusively endosperm. A second ROI will be a section that includes endosperm and, for black point, the brush end; a third ROI will be the entire grain as identified by a masking procedure. With each ROI, image processing will be done at the pixel level, whereby subregions of defect in the ROI are first identified; then, depending on the size of the subregion, a decision will be made on whether to categorize the ROI and/or the grain as normal or defective. Image analysis will also be used to develop a procedure for evaluating preharvest sprouting propensity in wheat breeding programs that can replace laborious and subjective methods of visually counting germinated seed and categorizing severity of sprouted seed within spikes.

Progress Report
Significant progress was made in the objectives of this project, which falls under National Program (NP) 306, Component 1, Problem Statement 1.A: Define, Measure, and Preserve/Enhance/Reduce Attributes that Impact Quality and Marketability, for which the NP 306 Action Plan for 2020-2024 (USDA-ARS-ONP 2019) states that "NP 306 research will develop technologies that improve quality, extend product shelf life, reduce waste, and decrease costs through innovative processing and packaging." At the request of the cereals industry, research continued characterizing the sources of variation in the wheat quality procedure known as ‘falling number’. This is a worldwide procedure that measures the viscosity of starch-gelatinized wheat meal under a prescribed heating regime. In that process indirectly characterizes the activity of the primary enzyme that digests starch, hence the integrity of the grain for the production of various wheat-based foods. Excessive enzyme activity translates into poor product quality and is usually associated with adverse weather before harvest. In the past year, a study was conducted to examine the effect of dissolved gases in the purified water used in the test tubes containing wheatmeal. Past observations have shown that while the water: meal mixture is heated and eventually brought to boiling temperature, voids form within the mixture, and their random formation causes variation in the speed of the viscometer’s stirring rods that fall through the mixture under the force of gravity after an agitation period. Three possibilities may cause the voids: air-entrained during agitation, steam bubbles formed after the mixture reaches boiling temperature, and dissolved gases that become released during heating. The study examined the third cause. By evacuating (by vacuum) a volume of distilled deionized water, the water was then used in the standard mixture formulation (7 g meal + 25 mL water) placed in a precision test tube, and immersed in boiling water. Results indicate a slight improvement in the precision of the falling number procedure when the water is degassed; however, the practical benefit is limited, so the dominant source of variation in the procedure must be attributed to one or both of the other factors. Other research in the past year has sought to better understand the potential for multispectral and hyperspectral imaging of wheat kernels for the tradeoffs between the breadth of information contained in full spectral images and the speed of image processing operation, with simplified processing translating to a faster speed. For example, seeds of cereal grains may undergo a post-harvest sorting operation for the removal of foreign material or physiological damage. Existing commercial sorters typically use the capture of light reflected from the object surface for accepting or rejecting seed, with an emphasis placed on speed of operation. Advances in spectral imaging and computation make it feasible to perform rudimentary image (1-3 wavebands) analysis in seed sorting operations. A study was conducted to examine the effect of size of the region of a seed undergoing image analysis, with size ranging from the entire viewed surface down to a small centrally located subregion (approximately 5 percent) of the surface. The objective was to determine how small size can be used while still maintaining the ability to categorize wheat seed into an ‘accept’ or ‘reject’ condition for sorting purposes. In a test case involving sound wheat kernels and kernels infected with Fusarium head blight (a fungal disease), it was found that image subregions representing less than 10 percent of the exposed seed surface produced reasonably accurate classifications, thus allowing for simplified image classification algorithm design. This research is intended to benefit developers of multispectral and hyperspectral image software for use in image-based cereal sorters and will ultimately lead to the adoption of image analysis in cereal seed sorting.

Accomplishments
1. Repeatability and reproducibility precision of a proposed standard reference material for the falling number method. Falling Number is a method used the world over to gauge the quality of harvested wheat, specifically the integrity of the starchy endosperm from which most wheat products are derived. It is a physical (starch pasting viscosity) technique that is an indirect indicator of enzymatic stability. Compared to the majority of cereals quality analytical procedures, falling Number has excellent precision; however, there is a need to use a stable starch or starch-like material as a standard reference material (SRM) to ensure single instruments, as well as devices in a network, fall within expected levels of precision and accuracy. USDA-ARS scientists at Beltsville, Maryland, in cooperation with USDA-FGIS laboratories, conducted a study that determined the precision of the falling number method using native corn starch as a SRM. Both intra- and inter-laboratory components of precision were determined. The establishment of these components will specify acceptable limits for control chart operations in a network of instruments in federal, state, and private consortiums .

Review Publications
Delwiche, S.R., Baek, I., Kim, M.S. 2021. Effect of curvature on hyperspectral reflectance images of cereal seed-sized objects. Biosystems Engineering. 202: 55-65. https://doi.org/10.1016/j.biosystemseng.2020.11.004.
Delwiche, S.R., Liang, J. 2020. On the use of native corn starch as a standard reference material for falling Number. Cereal Chemistry. 97(6):1227-1235. https://doi.org/10.1002/cche.10346.
Sim, E., Park, E., Ma, F., Baik, B.V., Fonseca, J.M., Delwiche, S.R. 2020. Sensory and physicochemical properties of whole wheat salted noodles under different preparations of bran. Journal of Cereal Science. 96(1):Article 103112. https://doi.org/10.1016/j.jcs.2020.103112.