A mosaic monoploid reference sequence for the highly complex genome of sugarcane

Sugarcane (Saccharum spp.) is a major crop for sugar and bioenergy production. Its highly polyploid, aneuploid, heterozygous, and interspecific genome poses major challenges for producing a reference sequence. We exploited colinearity with sorghum to produce a BAC-based monoploid genome sequence of sugarcane. A minimum tiling path of 4660 sugarcane BAC that best covers the gene-rich part of the sorghum genome was selected based on whole-genome profiling, sequenced, and assembled in a 382-Mb single tiling path of a high-quality sequence. A total of 25,316 protein-coding gene models are predicted, 17% of which display no colinearity with their sorghum orthologs. We show that the two species, S. officinarum and S. spontaneum, involved in modern cultivars differ by their transposable elements and by a few large chromosomal rearrangements, explaining their distinct genome size and distinct basic chromosome numbers while also suggesting that polyploidization arose in both lineages after their divergence.