Genotyping-by-Sequencing Derived Single Nucleotide Polymorphisms Provide the First Well-Resolved Phylogeny for the Genus Triticum (Poaceae)

Abstract

Wheat (Triticum spp.) is an important staple food crop for mankind since the beginning of agriculture. The genus Triticum L. is composed of diploid, tetraploid and hexaploid species in which majority of the species have not yet been discriminated clearly and hence their phylogeny and classification remains unresolved. Genotyping-by-sequencing (GBS) is an easy and affordable method that allows us to generate genome-wide single nucleotide polymorphism (SNPs) markers. In this study, we used GBS to obtain SNPs covering all seven chromosomes from 283 accessions of Triticum related genera. After filtering low quality and redundant SNPs based on haplotype information, the GBS assay provided 14,188 high-quality SNPs that were distributed across the A (71%), B (26%), and D (2.4%) genomes. Cluster analysis and discriminant analysis of principal components (DAPC) allowed us to distinguish six distinct groups that matched well with Triticum species complexity. We constructed Bayesian phylogenetic tree using 14,188 SNPs, in which 17 Triticum species and subspecies have discriminated. The dendrogram analysis revealed that the polyploid wheat species could be divided into groups according to the presence of A, B, D and G genomes with strong nodal support and provided a new insight on the evolution of spelt wheat. A total of 2,692 species specific SNPs were identified to discriminates the common (T. aestivum) and durum (T. turgidum) wheat cultivar and landraces. In PCoA grouping, the two wheat species formed individual cluster and the SNPs were able to distinguish up to nine groups of 10 subspecies. This study demonstrated that GBS derived SNPs could be used efficiently in genebank management to classify Triticum species and subspecies that are very difficult to distinguish by their morphological characters.