A Whole Genome Sorting-Whole Genome Shotgun (WGS2) Approach for Sequencing the Allotetraploid Durum Wheat Genome

The allotetraploid (AABB) durum wheat (Triticum turgidum) has a ≈11 Gb genome and is economically important for making pasta and certain types of bread. Like common wheat, its complexity turns whole genome sequencing and assembly using current mainstream short read-technologies into a discouraging task. The newer single-molecule long-read technologies would in theory allow resolving the genomes was it not that for a 11Gb genome the sequencing would be difficult and too costly. Using an updated strategy, combining technologies, we started sequencing the “pre-introgression breeding” durum wheat Senatore Cappelli, dated 1915, as the progenitor of the majority of modern pasta varieties.

Our strategy is based on bulk-separation of the A- and B-genome using Fluorescence In Situ Hybridization In Suspension (FISHIS), resolving the polyploidy-problem, and was used to prepare Illumina libraries for each genome. The sequencing itself has been done step-wise, allowing smooth shifting towards improving technologies. Furthermore, assemblies of grass genomes, due to repeat-content, hardly yields very large contigs. Therefore efforts have been made to generate long span mate-pairs and a whole genome optical map using high molecular weight gDNA. An assembly of the A genome has been obtained and is being annotated, guided by RNA-Seq data. Our aim is to obtain high quality (L50≥300 kb), annotated assemblies of both A and B genomes. These will become the template for re-sequencing modern varieties and wild relatives, enabling the reconstruction of the phylogeny of this important crop. We believe the WGS² approach described here will serve as an example for sequencing allopolyploid crops, given FISHIS labeling applies.