Illumina TruSeq Synthetic Long-Read Sequence Enhances de novo Genome Assembly of Allotetraploid White Clover (Trifolium repens L.)

De novo assembly of reference genomes remains a challenge for some economic species.  Polyploidy, a feature common to many plants, impedes assembly when reliant on short-read sequencing platforms. White clover, a prominent forage legume in grazed temperate pasture, has a relatively compact genome (≈1 Gb). It presents particular sequencing challenges as an allogamous allotetraploid (2n=4x=32) derived from closely related diploid progenitors.  Although a range of genomic resources have been developed including expressed sequence and methyl-filtered libraries, structured populations and linkage maps, a reference genome is only now emerging from a collaboration between Pastoral Genomics and Aarhus University.

A de novo sequencing strategy utilising an inbred (S₉) genotype and based primarily on Illumina short-read platforms (≈200x coverage) has been augmented by Illumina TruSeq Synthetic Long-Read sequencing.  Six Long-Read libraries were generated.  Library mean yield was 560 Mb of sequence data assembled into an average of 111,000 long-reads ≥ 1500 bp/library.  Sequence length ranged from 1500 bp to 14 Kb with a sequencing yield mode of 10 Kb.

Inspection of the augmented assembly indicated a range of improvements attributable to the Long-Read sequences.  These included validation of existing contigs/scaffolds derived from short-read sequence, correction of small gaps and mis-scaffolding, identification of assembly conflation between homoeologues, and significant large-gap and contig-end filling.

These findings suggest Long-Read sequence data offer a valuable tool to aid assembly of complex polyploid genomes, and have added considerably to the accuracy of the white clover reference genome assembly and its value for downstream applications in species understanding and improvement.