Calibration equations for the estimation of amino acid composition in whole soybeans were developed using partial least squares (PLS), artificial neural networks (ANN), and support vector machines (SVM) regression methods for five models of near-infrared (NIR) spectrometers. The effects of amino acid/protein correlation, calibration method, and type of spectrometer on predictive ability of the equations were analyzed. Validation of prediction models resulted in r 2 values from 0.04 (tryptophan) to 0.91 (leucine and lysine). Most of the models were usable for research purposes and sample screening. Concentrations of cysteine and tryptophan had no useful correlation with spectral information. Predictive ability of calibrations was dependent on the respective amino acid correlations to reference protein. Calibration samples with nontypical amino acid profiles relative to protein would be needed to overcome this limitation. The performance of PLS and SVM was significantly better than that of ANN. Choice of preferred modeling method was spectrometerdependent. Calibration equations for the estimation of amino acid composition in whole soybeans were developed using partial least squares (PLS), artificial neural networks (ANN), and support vector machines (SVM) regression methods for five models of near-infrared (NIR) spectrometers. The effects of amino acid/ protein correlation, calibration method, and type of spectrometer on predictive ability of the equations were analyzed. Validation of prediction models resulted in r 2 values from 0.04 (tryptophan) to 0.91 (leucine and lysine). Most of the models were usable for research purposes and sample screening.Concentrations of cysteine and tryptophan had no useful correlation with spectral information. Predictive ability of calibrations was dependent on the respective amino acid correlations to reference protein. Calibration samples with nontypical amino acid profiles relative to protein would be needed to overcome this limitation. The performance of PLS and SVM was significantly better than that of ANN. Choice of preferred modeling method was spectrometer-dependent.

A key element of successful development of new soybean cultivars is availability of inexpensive and rapid methods for measurement of FA in seeds. Published research demonstrated applicability of NIR spectroscopy for FA profiling in oilseeds. The objectives of this study were to investigate the applicability of NIR spectroscopy for measurement of FA in whole soybeans and compare performance of calibration methods. Equations were developed using partial least squares (PLS), artificial neural networks (ANN), and support vector machines (SVM) regression methods. Validation results demonstrated that (i) equations for total saturates had the highest predictive ability (r 2 = 0.91-0.94) and were usable for quality assurance applications, (ii) palmitic acid models (r 2 = 0.80-0.84) were usable for certain research applications, and (iii) equations for stearic (r 2 = 0.49-0.68), oleic (r 2 = 0.76-0.81), linoleic (r 2 = 0.73-0.76), and linolenic (r 2 = 0.67-0.74) acids could be used for sample screening. The SVM models produced significantly more accurate predictions than those developed with PLS. ANN calibrations were not different from the other two methods. Reduction in the number of calibration samples reduced predictive ability of all equations. The rate of performance degradation of SVM models with sample reduction was the lowest. a r 2 is determination coefficient, SEP is SE of prediction corrected for bias, d is bias, RPD is relative predictive determinant, PLS is partial least squares, ANN is artifical neural networks, and LS-SVM is Least Squares support vector machines. b Model parameters provide number of latent variables for PLS, number of inputs and neurons in a hidden layer for ANN, and radial basis function bandwidth and complexity regularization parameter for LS-SVM. NIR MEASUREMENT OF SOYBEAN FATTY ACIDS 425 JAOCS, Vol. 83, no. 5 (2006) FIG. 2. Actual vs. predicted concentration plots for saturates and linolenic FA calibrations. Models were tested on sets of 180 (saturates) and 244 (linolenic) samples. The solid line on each plot represents the regression line.

No abstract

Rheological properties of raw oat flour slurries were determined in experimental high β‐glucan (≤7.8%) and traditional oat lines (4–5% β‐glucan) grown in two consecutive years. Three different media were used to disperse oat flours: deionized water, silver nitrate solution (to inactivate endogenous enzymes), and alkali solution (to solubilize both water‐soluble and water‐insoluble β‐glucans). Significant correlations (P < 0.05) between viscosity of slurries and β‐glucan concentration obtained in either deionized water (r = 0.833), silver nitrate (r = 0.940), or alkali (r = 0.896) solutions showed that β‐glucans were the main contributor to oat extract viscosity. The highest correlation was obtained in silver nitrate solution, suggesting that inactivating endogenous enzymes is important to obtain high correlations. Predictive models of oat β‐glucan concentration based on the viscosity profile were developed using partial least squares (PLS) regression. Prediction of β‐glucan concentration based on viscosity was most effective in the silver nitrate solution (r = 0.949, correlation coefficient of predicted vs. analyzed β‐glucans) and least effective in the alkali solution (r = 0.870). These findings demonstrate that the β‐glucan in oat could be predicted by measuring the viscosity of raw flours in silver nitrate solution, and this method could be used as a screening tool for selective breeding.

The vane method was applied to evaluate failure characteristics of soy‐based yogurts prepared from five soybean varieties at Brix values of 6, 8, and 10°. Yield stress, yield strain, and water‐holding capacity were compared. Yield stress values ranging from 133 to 420 Pa at 2.5% protein and 498 to 1171 Pa at 4.0% protein were dependent on soybean variety and increased with increasing protein concentration. The average yield strain of samples was not affected by protein or variety. Compared to commercial dairy yogurt, soy yogurt had 132 to 445% higher yield stress at similar protein content, and was less deformable based on yield strain measurements. Water‐holding capacity of soy yogurts was variety dependent, although this dependence was less pronounced at higher protein concentrations. The vane method may be effectively used as a rapid and inexpensive technique for detecting textural differences of soy‐based yogurts.