Project description:Polyploidy poses challenges for phylogenetic reconstruction because of the need to identify and distinguish between homoeologous loci. This can be addressed by use of low copy nuclear markers. Panicum s.s. is a genus of about 100 species in the grass tribe Paniceae, subfamily Panicoideae, and is divided into five sections. Many of the species are known to be polyploids. The most well-known of the Panicum polyploids are switchgrass (Panicum virgatum) and common or Proso millet (P. miliaceum). Switchgrass is in section Virgata, along with P. tricholaenoides, P. amarum, and P. amarulum, whereas P. miliaceum is in sect. Panicum. We have generated sequence data from five low copy nuclear loci and two chloroplast loci and have clarified the origin of P. virgatum. We find that all members of sects. Virgata and Urvilleana are the result of diversification after a single allopolyploidy event. The closest diploid relatives of switchgrass are in sect. Rudgeana, native to Central and South America. Within sections Virgata and Urvilleana, P. tricholaenoides is sister to the remaining species. Panicum racemosum and P. urvilleanum form a clade, which may be sister to P. chloroleucum. Panicum amarum, P. amarulum, and the lowland and upland ecotypes of P. virgatum together form a clade, within which relationships are complex. Hexaploid and octoploid plants are likely allopolyploids, with P. amarum and P. amarulum sharing genomes with P. virgatum. Octoploid P. virgatum plants are formed via hybridization between disparate tetraploids. We show that polyploidy precedes diversification in a complex set of polyploids; our data thus suggest that polyploidy could provide the raw material for diversification. In addition, we show two rounds of allopolyploidization in the ancestry of switchgrass, and identify additional species that may be part of its broader gene pool. This may be relevant for development of the crop for biofuels.
Project description:Switchgrass (Panicum virgatum) is a polyploid, outcrossing grass species native to North America and has recently been recognized as a potential biofuel feedstock crop. Significant phenotypic variation including ploidy is present across the two primary ecotypes of switchgrass, referred to as upland and lowland switchgrass. The tetraploid switchgrass genome is approximately 1400 Mbp, split between two subgenomes, with significant repetitive sequence content limiting the efficiency of re-sequencing approaches for determining genome diversity. To characterize genetic diversity in upland and lowland switchgrass as a first step in linking genotype to phenotype, we designed an exome capture probe set based on transcript assemblies that represent approximately 50 Mb of annotated switchgrass exome sequences. We then evaluated and optimized the probe set using solid phase comparative genome hybridization and liquid phase exome capture followed by next-generation sequencing. Using the optimized probe set, we assessed variation in the exomes of eight switchgrass genotypes representing tetraploid lowland and octoploid upland cultivars to benchmark our exome capture probe set design. We identified ample variation in the switchgrass genome including 1,395,501 single nucleotide polymorphisms (SNPs), 8173 putative copy number variants and 3336 presence/absence variants. While the majority of the SNPs (84%) detected was bi-allelic, a substantial number was tri-allelic with limited occurrence of tetra-allelic polymorphisms consistent with the heterozygous and polyploid nature of the switchgrass genome. Collectively, these data demonstrate the efficacy of exome capture for discovery of genome variation in a polyploid species with a large, repetitive and heterozygous genome.