One Plant - Multiple Genomes: Phasing Homeologs in Polyploid Wheat

Together with rice and corn, wheat provides a major source of food to human population. However, the genomic tools in wheat are limited due to the challenges posed by polyploidy and large genome size. Cultivated wheat is derived from two main species, tetraploid Triticum turgidum (AABB genome, 12 Gb) and hexaploid Triticum aestivum (AABBDD genome, 18 Gb). Despite the large size of the genome, gene-coding region of wheat is comparable to other grasses and constitutes only 1-2% percent of genomic sequence. To expand wheat genomic resources, we have sequenced and assembled transcriptomes of a tetraploid Triticum turgidum and its closely related diploid relative Triticum urartu. We have examined several parameters affecting the assembly of polyploid wheat compared to the diploid wheat and incorporated additional steps necessary for the genome-specific assembly. With our assembly strategy, 80% of A and B homeologs are already separated in different contigs as estimated from benchmark A/B sequences. However, a significant proportion of sequences represent merged and often chimeric copies. We have developed a pipeline for identification and re-assembly of the merged contigs that consists of the following steps: 1) re-mapping all reads back to the assembly 2) identifying polymorphisms 3) phasing polymorphisms 4) performing local re-assembly 5) evaluating the result against the diploid T. urartu assembly. Resulting wheat gene sets are available from the project website: http://maswheat.ucdavis.edu/Transcriptome/index.htm