Mining Genetic Variance in Sugarcane through Genotyping By Sequencing

Sugarcane (Saccharum spp.)  is one of the most important economic crops for its high sugar and biofuel production. Investigating genomic sequence variances is critical for identifying alleles contributing to important agronomic traits. However, the genomic polymorphism detection in sugarcane is lagging behind due to its highly polyploid and complex genomes. To mine genetic variance in sugarcane, we applied genotyping by sequencing (GBS) on 14 representative germplasm accessions. Different strategies for genetic variants calling were compared and an optimal pipeline for polyploid sugarcane was established. A total of 76,251 non-redundant single nucleotide polymorphisms (SNPs), 5,642 InDels, and 4,236 presence/absence variants were detected among the 14 accessions. More than 90% of SNPs were bi-allelic and about 40% of them are single dose alleles. The percentages of homozygous loci of each accession were estimated in a range of 72% to 91% based on high confidence SNPs. The efficiency to detect polymorphism in sugarcane was compared between two complex reduction sequencing methods, GBS and target enrichment sequencing (TGS). The density of SNPs detected by TGS was higher than that by GBS. However, the GBS has the potential to detect increased number of SNPs if the sequencing depth is increased. The results demonstrated that GBS provided an effective way to discover genomic sequence variation in highly polyploid and heterozygous species, which will facilitate genetic and genomic studies for polyploidy sugarcane improvement.