Use of student’s t statistic as a phenotype of relative consumption preference of wheat (Triticum aestivum L.) grain

Authors: Kiszonas, Alecia; FUERST, E - Washington State University; Morris, Craig

Submitted to: Journal of Cereal Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/17/2015
Publication Date: 9/1/2015
Publication URL: http://handle.nal.usda.gov/10113/61325
Citation: Kiszonas, A., Fuerst, E.P., Morris, C.F. 2015. Use of student’s t statistic as a phenotype of relative consumption preference of wheat (Triticum aestivum L.) grain. Journal of Cereal Science. 65:285-289.

Interpretive Summary: A potential breeding objective of wheat (Triticum spp. L.) improvement is to make the grain more palatable to humans and other mammals. Our present aim is to evaluate the use of a common check variety to provide a quantitative phenotype that can be used to identify QTL in association mapping studies. QTL would be for flavor or aroma compounds that contribute to differences in consumption preference. The use of check varieties for each sample variety allows for a unitless t statistic by which the varieties can be ranked and compared. The benefit of the t statistic is that it accounts for variance and sample size. A drawback to the commonly used preference or consumption ratios is that they are relative, non-independent measures, whereas the t statistic has clear direction. Our prior studies with mice used the tournament bracket design as is often seen in athletics, which is the best method when the objective is identifying the “Yummiest” and “yuckiest” variety in a group of samples. This style of experimental design, however, does not allow an objective ranking of the varieties. The t statistic, when all sample varieties are compared to common check varieties, does facilitate the objective ranking of the varieties within the sample varieties. In addition to our own ability to assess varieties based on these t statistics, they provide objective values that can be used as phenotypes in genetic mapping in an effort to gain a better understanding of the genes/loci behind flavor and food preference.

Technical Abstract: Whole-grain wheat (Triticum aestivum L.) products provide essential nutrients to humans, but bran attributes may hinder consumption. Differences in grain attributes including flabor/aroma can be indentified using the house mouse (Mus musculus L.) as a model system. A potential application of this model system is to identify genes or quantitative trait loci (QTLs) conferring consumption-related traits using genetic mapping. These analyses require a quantitative phenotype to test marker-trait associations. Two-choice feeding trials are common in consumption studies, but do not provide an independent, quantitative phenotype. The objective of this study was to examine the use of ‘check’ varieties against which to compare sample varieties to generate Student’s t values for use as phenotypic values. Two checks, which had previously been identified as “Yummy” and “yucky” were compared against each sample variety. The resulting t values provided an independent quantitative phenotype of the sample varieties.