Submitted to: Euphytica
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/29/2005
Publication Date: 8/29/2005
Citation: Fufa, H., Baenziger, P.S., Beecher, B.S., Dweikat, I., Graybosch, R.A., Eskridge, K.M. 2005. Comparison of phenotypic and molecular marker-based classifications of hard red winter wheat cultivars. Euphytica 133-146.
Interpretive Summary: For several decades, concern has been expressed over a possible loss of genetic diversity amongst cultivated wheats. Low levels of genetic diversity could leave the U.S. winter wheat crop susceptible to newly evolved races of pests and pathogens. Historically, genetic diversity has been estimated by use of pedigree analysis. However, pedigree analysis relies on hypothetical assumptions based on the derivation of new lines, and is not supported by direct empirical evidence. In this study, DNA, grain storage protein and phenotypic (overt characteristics) traits were used to estimate genetic diversity among winter wheat cultivars. The various measures differed in their diversity estimates, but all were lower than those observed by pedigree evaluation. This study suggests genetic erosion in U.S. wheat is more severe than suspected, and, in the future, direct measures of genetic diversity, rather than hypothetical ones, are necessary.
Technical Abstract: Genetic diversity is the basis for successful crop improvement and can be estimated by different methods. The objectives of this study were to estimate the genetic diversity of thirty ancestral to modern hard red winter wheat (Triticum aestivum L.) cultivars adapted to the Northern Great Plains using pedigree information, morphological traits (agronomic measurements from six environments), end-use quality traits (micro-quality assays on 50 g grain or milled flour samples for the six environments), and molecular markers (seed storage proteins separated using SDS-PAGE, 51 SSRs, and 23 SRAP DNA markers), and to determine the relationships of genetic distance estimates obtained from these methods. Relationships among diversity estimates were determined using simple (Pearson) and rank (Spearman) correlation coefficients between distance estimates and by clustering cultivars using genetic-distances for different traits. All methods found a wide range in genetic diversity. The genetic distance estimates based on pedigree had the highest values due to possible over-estimation arising from model assumptions. The genetic diversity estimates based on seed storage protein were lowest because they were the major determinants of end-use quality, which is a highly selected trait. In general, the diversity estimates from each of the methods were positively correlated at a low level with the exceptions of SRAP diversity estimates being independent of morphologic traits (simple correlation), SDS-PAGE, and SSR diversity estimates (rank correlation). However, SSR markers, thought to be among the most efficient markers for estimating genetic diversity, were most highly correlated with seed storage proteins. The procedures used to accurately estimate genetic diversity will depend largely upon the tools available to the researcher and their application to the breeding scheme.