Genetic diversity, population structure and association mapping of biofuel traits in southern switchgrass germplasm
Acharya, Ananta Raj
MetadataShow full item record
Switchgrass (Panicum virgatum L.), a warm season grass native to North America, is being developed as a biofuel crop. Plant breeding can improve biofuel characteristics further, particularly if the genetic diversity of germplasm resources is clearly understood. The objective of this study was to examine the population structure and relatedness within and among forty-nine switchgrass populations mainly derived from the southern United States and use the information to identify putative QTLs associated with biomass yield, plant height, stem diameter and days to flower. These populations included both upland and lowland ecotypes. A total of 511 genotypes were selected for genotyping and phenotyping. SSR markers developed from switchgrass and well distributed across the switchgrass genome were used to genotype the individuals. We used 35 markers and 365 alleles were discovered for those markers. In addition, we used a Genotyping-by-Sequencing (GBS) protocol to identify and utilize SNPs as genetic markers. With GBS, we identified 65,328 SNP markers. We only used 3,196 SNPs for our analysis, after filtering for read depth of at least 6 reads per locus per genotype and requiring no more than 20% missing genotypic data for any given locus. In order to investigate the effect of missing data, we also used a second dataset of about 20,000 SNPs allowing up to 50% of individuals to have missing genotypic data for any given locus. We also used nine chloroplast specific markers to identify the cytotype. The data were used to examine the population structure and to perform phylogenetic analysis. Along with measuring dry biomass after harvest, we collected three canonical morphological data; plant height, stem diameter and flowering time on the individuals. We found a population differentiation in the two major groups, upland and lowland ecotypes, with phenotypic, cytotypic and genotypic data. A deeper sub-population structure was identified within the broad lowland and upland population. The sub population structure was correlated with the geographical origins of those accessions. We were able to identify two groups within lowland ecotypes, one of which did not exhibit the typical morphological characteristics of lowland accessions. We also studied the association of markers with above mentioned traits, and within the limitations of the number of environments used, identified several QTL significantly associated with each trait.