Location: Market Quality and Handling Research
Title: Utilizing NIR to Predict Total Oil, Sugars, Fatty Acids and Tocopherols in Whole Peanut Seeds. Authors
Submitted to: American Peanut Research and Education Society Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: June 18, 2012
Publication Date: N/A
Interpretive Summary: Methods are needed that are quick, cheap and do not destroy the sample are needed to determine the nutritional composition of peanuts by breeders in order to produce varieties with improved nutritional characteristics. Near Infrared Spectroscopy (NIR) was evaluated for the fatty acid, palmitic acid, total oil content, vitamin E content and total sugar content. The samples evaluated were from the 2009 and 2010 plots used for the Uniform Peanut Performance Tests (UPPT) which represent the Southern US peanut growing regions. Comparisons were done with the data obtained by the current methods of instrumentation. It was found that the results for total oil and fatty acid content were more accurately measured by NIR than vitamin E and sugars. Although accuracy was lacking, the speed and cost of the NIR was acceptable when compared with having no data to use.
Technical Abstract: Improving nutritional composition of peanut is a goal of many peanut breeding programs. Among the factors under consideration are total oil, fatty acids, antioxidants, and sugars. Because breeding programs develop and evaluate thousands of segregating progeny each year, they require rapid, inexpensive, and often non-destructive methods to evaluate chemical composition of peanuts. One such tool is Near Infrared Spectroscopy (NIR).The University of Florida peanut breeding program has utilized a ThermoNicolet Nexus 670 FT-IR scanning monochronometer equipped with a NearIR UpDrift Smart to develop predictions for oleic and linoleic fatty acids. This report will summarize progress in developing NIR predictions for palmitic fatty acid, total oil, tocopherols and sugars. Total oil, tocopherols and sugars were measured from samples harvested in 2009 and 2010 from field plots across the entire peanut growing region of the southern United States (Alabama, Florida, Georgia, North Carolina, Oklahoma, South Carolina, Texas and Virginia as part of the Uniform Peanut Performance Tests (UPPT). Constituents were measured by the USDA-ARS Market Handling and Quality Research Unit (MHQRU). Spectra were obtained using a subsample of the seeds used for testing by MHQRU. Calibration models were developed for total sugars, oils and tocopherols. Separately, spectral data were obtained on single seeds and palmitic acid was measured in the seeds by Gas Chromatography. Preliminary NIR predictions utilizing whole, single seeds were developed for total oil, palmitic acid, total tocopherols and total sugars. The calibration equation predicted palmitic acid with an R2 of 0.72 and a slope of 0.95. Five individual seeds from each of 19 cultivars were used to externally validate the calibration equation. NIR predicted palmitic acid from the external validation set with an R2 of 0.80 and a slope of 1.04. Calibration results for total oil showed R2 of 0.84. Results show that prediction of total tocopherols (R2=0.65) and sugars (R2=0.71) are not as precise as total oil and oleic/linoliec/palmitic fatty acids. Current methodologies for measuring tocopherols and sugars are not used routinely in peanut breeding programs because they are slow and expensive. Therefore, even a moderate level of predictive accuracy is preferable to no information.