2013 Annual Report
1a.Objectives (from AD-416):
1) Update calibrations for grain composition and amino acids,.
2)update calibrations for and standardize five cooperator instruments, and.
3)train and support cooperators in quality control and lab operations.
1b.Approach (from AD-416):
The calibration samples will be provided by collaborating plant breeders. We estimate that 400 total samples per year will be screened for possible use as calibration samples. The GCL NIRS values will be compared to any original values from the breeder cooperator, which will provide an estimate of testing variability (standard deviation). GQL will scan the samples, and will collect the spectral data, sample information, and measured values. Samples for reference analysis (n=100; 50, 50, 50 per year, respectively) will be sent to the University of Missouri. Calibrations will be validated annually according to AACC Method 39-00, and updated if needed by April 30 of the respective marketing year. Updates are done by inclusion of the new samples in the calibration data set of the machine brand being evaluated. Calibrations/instrument setup for the following fall harvest will be done by the start of harvest. Temperature stability is introduced/verified by testing the 10-annual temperature samples at four temperatures (-5, 5, 25, and 45oC), using the accuracy criteria of AACC Method 39-00 to determine if additional temperature stability is needed. Standardization/traceability to the source of calibrations is necessary for accurate data from users’ instruments. Five cooperator instruments will be standardized against the calibration master units at GQL and against the wet chemistry references from University of Missouri, using 40 corn samples that have at least 3 replicate reference values for each factor. Annual verification of the standardization will be done with data from the cooperator samples. A new standardization with the GQL master samples will be needed if the calibrations are changed during the project period. The presence of these verified cooperator units will form the basis for a network of testing sites for cooperators specializing in marketing or utilizing the kinds of organic grains we are breeding. Project cooperators will determine the ongoing operating structure of this network, if created. Initial training, then annual support will be held with cooperators (at the time of the annual project meeting), with the intent to have cooperators self-sufficient in quality control and lab operations before the end of the project.
Iowa State University (ISU) continued to improve their Near-Infrared Spectroscopy (NIRS) calibrations for the essential amino acids lysine, methionine and cysteine and protein, oil and starch by sending poorly predicted and/or spectrally significant samples to the Experiment Station Chemistry Laboratory at the University of Missouri for laboratory analysis. The resulting data were used to develop new calibration equations. These calibration equations were provided to ARS and installed on the ARS NIR instrument. A parallel set of calibrations for another type of analyzers is also being developed. Calibration samples originated from collaborators in the 2006-2010 crops. The standard errors of cross validation were 0.021, 0.017, and 0.023 percentage points dry basis, for lysine, methionine, and cysteine, respectively. Samples from the 2011 crop and 2012 crop were held out for independent validation in July 2013.
Calibrations were installed in one ARS unit and three ISU units on campus.
These calibrations improve the ability of NIRS to identify useful material in breeding populations. Useful in this context means with essential amino acid profiles above the normal profile correlated with the corn protein content. Hybrids with improved profiles will be more valuable for animal or human nutrition. The independent validation and next update, to take place in July 2013, will be accompanied by an economic evaluation of costs for calibration and operation versus potential feed value benefits of improved composition.
Because the calibration transfer process for the amino acid calibrations was not possible outside Iowa State University, the project focus shifted to become an update of the full set of proximate analysis calibrations for the same units as well as the amino acid calibrations. The samples provided by the organic breeder collaborators were at the extremes of proximate analysis as well as being unique in amino acid profiles. The result by harvest 2013 will be an expanded-range set of corn proximate analysis calibrations and a validated set of amino acid calibrations. The major question in 2013-2014 will be how to distribute these calibrations with a quality control and validation protocol so that organic breeders will have low cost methods for improving quality of their grains. We understand that the calibrations will also be applicable to commodity corn hybrids as well; however, the diversity of material encountered by organic breeders indicates that the organic community will gain significant benefits from the ability to do rapid analysis.