Ploidy estimates with genome-wide heterozygosity
For GBS-genotyped samples that vary in ploidy, measures of genome-wide
heterozygosity can potentially be used as a complement to flow cytometry
for identifying ploidy variation (Gompert & Mock, 2017). This approach
is based on the rationale that if all reads are mapped to a diploid
reference genome, any subgenome differences in polyploids will be
recorded as ‘heterozygous’ loci. At increasing ploidy levels, each
additional subgenome will introduce more differences and increase the
number of heterozygous locus calls. This approach can also be used to
identify diploid hybrids, which are identifiable as showing a quantum
increase in heterozygosity with no associated increase in genome size
based on flow cytometry. We employed these strategies to assess ploidy
variation in the Paspalum samples, focusing on genotypically
unique accessions with the highest sequence coverage. Genome-wide
heterozygosity was calculated as the individual inbreeding coefficient
(F) with the het function in plink (Purcell et al., 2007).