Submitted to: Cereal Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/1/1998
Publication Date: N/A
Citation: N/A Interpretive Summary: Wheat-rye (Triticum aestivum L - Secale cereale L) chromosomal translocations (designated as 1AL.1RS or 1BL.1RS), particularly those involving the short arm of rye chromosome 1R, have been used during the past 20 years to instill resistance to plant pathogens and insects, and improve the hardiness, adaptation, and yield of wheat. Unfortunately, the presence of the 1AL.1RS or 1BL.1RS rye translocations in wheat has been shown to impart inferior dough handling and baking characteristics. Although numerous analytical techniques (e.g., HPLC, monoclonal antibody tests, high performance capillary electrophoresis) have been developed for detecting these translocations, the complexity of the analytical procedures restricts their use to research and analytical laboratories. The purpose of this study was to examine the potential of diffuse reflectance near-infrared (NIR) spectroscopy, a well accepted technique in the grain industry, for detection of 1RS in wheat samples. By use of a regression procedure that is common in NIR spectroscopy, we were able to develop classification models that were 99-100% accurate when classifying commercial wheats. Accuracy dropped to 80-85% when wheat samples originated from plant breeders stock because of the closeness in genetic makeup. This work is of potential benefit to the wheat trade and processing industries.
Technical Abstract: Detection of wheat-rye (Triticum aestivum L - Secale cereale L) 1AL.1RS and 1BL.1RS chromosomal translocations by diffuse reflectance near-infrared (NIR) spectroscopy was examined. Such translocations are imparted in wheat for the purpose of improving resistance to plant pathogens and insects, and improving yield and hardiness. However, deleterious effects occur as a result of the replacement of wheat gliadins with rye secalins, notably poorer baking quality. Three independent groups of wheat samples, ranging in genetic diversity from sister lines derived from 1RS breeding populations to commercial cultivars, were studied. NIR spectra (1100-2498 nm) of wheat flour samples within each group were modeled using partial least squares (PLS) regression by assigning a value of 0 or 1 to each sample, with the choice of value dependent on the absence or presence of the 1RS translocation. Using a 0.5 midpoint value as the criterion for deciding whether a sample was correctly classified, one-sample-out cross-validation demonstrated classification accuracy rates as high as 100%, particularly with commercial cultivars. When applied to a separate set of samples within the same group, accuracy was still very high (ca. 99%). However, accuracy dropped to 80-85% when samples of more closely related genetic backgrounds were modeled. Most problematic are samples that are heterogeneous for 1RS, such as the cultivar Rawhide. Incorporating heterogeneous samples into a calibration results in an improvement in classification accuracy for these samples, but may diminish the prediction accuracy for non-heterogeneous samples.