Genetic Variation and Relationships of Pedigree-Known Oat, Wheat, and Barley Cultivars Releaved by Bulking and Single-Plant Sampling

Applications of bulking procedures have played an increasingly important role in molecular characterization of plant germplasm, but little attention has been made to address the effectiveness of detecting genetic variation and inferring genetic relationships via bulking. An analysis was performed here to compare the genetic variation detected and genetic relationships inferred via bulking and single-plant sampling of five oat (Avena sativa L.), wheat (Triticum aestivum L.), and barley (Hordeum vulgare L.) cultivars with known pedigrees using amplified fragment length polymorphism (AFLP) markers. Three AFLP primer pairs were applied to screen one bulk and eight single-plant samples of each cultivar and up to 140 AFLP bands were scored for each sample. Analyses of these AFLP data showed bulking revealed AFLP variation up to 21.4% less than corresponding single-plant sampling for these crop species and also introduced up to 2.2% upward and 5.1% downward biases in detecting AFLP variations for each cultivar. The genetic relationships inferred by bulking using the Dice's coefficient, the simple matching coefficient, and the Jaccard's coefficient were largely the same, but differed from those in single-plant samples employing average Dice's coefficient, average simple matching coefficient, and AMOVA-based distance method. All of the inferred genetic relationships were not congruent to the known pedigrees. Clearly, substantial biases could exist in detection of AFLP variation and in inference of genetic relationships from bulk samples, even for closely related germplasm, and more efforts to assess the effectiveness of bulking in inferring genetic variation and relationships are needed for more informative molecular characterization of plant germplasm.