BioNano genome mapping of individual chromosomes supports physical mapping and sequence assembly in complex plant genomes

Abstract

The assembly of a reference genome sequence of bread wheat is challenging due to its specific features such as the genome size of 17 Gbp, polyploid nature and prevalence of repetitive sequences. BAC-by-BAC sequencing based on chromosomal physical maps, adopted by the International Wheat Genome Sequencing Consortium as the key strategy, reduces problems caused by the genome complexity and polyploidy, but the repeat content still hampers the sequence assembly. Availability of a high-resolution genomic map to guide sequence scaffolding and validate physical map and sequence assemblies would be highly beneficial to obtaining an accurate and complete genome sequence. Here, we chose the short arm of chromosome 7D (7DS) as a model to demonstrate for the first time that it is possible to couple chromosome flow sorting with genome mapping in nanochannel arrays and create a de novo genome map of a wheat chromosome. We constructed a high-resolution chromosome map composed of 371 contigs with an N50 of 1.3 Mb. Long DNA molecules achieved by our approach facilitated chromosome-scale analysis of repetitive sequences and revealed a ~800-kb array of tandem repeats intractable to current DNA sequencing technologies. Anchoring 7DS sequence assemblies obtained by clone-by-clone sequencing to the 7DS genome map provided a valuable tool to improve the BAC-contig physical map and validate sequence assembly on a chromosome-arm scale. Our results indicate that creating genome maps for the whole wheat genome in a chromosome-by-chromosome manner is feasible and that they will be an affordable tool to support the production of improved pseudomolecules.

Keywords: chromosomes; flow sorting; optical mapping; physical map; sequencing; wheat.

Figure 1

Genome map No. 350 comprising a long array of tandem repeats. The pile of single molecules, depicted as yellow lines with blue and green dots corresponding to mapped and unmapped labels, respectively, was a source for building the consensus genome map (blue bar). The regular labelling pattern indicates presence of tandem repeats.

Figure 2

Scaffolding and correcting physical map contigs based on the genome map No. 19. (a) In total, nine contigs of the physical map (black bars) could be anchored through sequences of constituting BAC clones (blue lines) to the genome map No. 19 (green bar). Short red bars indicate approximate positions of Aegilops tauschii SNP markers anchored to particular clones. The purple line and bar in ctg3865 represent clone TaaCsp067A19, which was incorrectly assigned to this contig. Detail is shown in (b). A cmap of the clone TaaCsp067A19 does not match the corresponding region in GM19. The green bar corresponds to GM19, while the blue bars represent in silico digested BAC sequences (cmaps).

Figure 3

Local integration of three genetic maps through the genome map No. 15. Contigs of the physical map (black bars) were aligned to the genome map No. 15 (green bar) through sequences of constituting BAC clones (blue bars). The BAC contigs carry genetic markers (red) originating from three genetic maps, which could be integrated through the genome map.

Figure 4

Merging genome maps. Three genome maps (green bars) could be merged together after aligning sequences of BAC clones (blue bars) from three contigs of the physical map (black bars).