Construction of Comprehensive Reference Transcripts from Three Homoeologous Genomes in Common Wheat

Common wheat is an allohexaploid (genome formula: AABBDD), with huge and complex genome, consisting of approximately 80% repetitive sequences. Currently, international projects on the genome sequencing of common wheat are ongoing. Hence, we collected approximately 700,000 pair-end ESTs and 18,000 complete full-length cDNAs sequences constructed using the cap-trapper method. For annotation of expressed genes in the genome, we assembled ESTs to construct reference transcript sets for common wheat. Wheat ESTs were assembled using Mira. Because consensus SNPs were conserved among homoeologs, contents of the chimeric assemblies among homoeologous group genes were less than typically expected. We then clustered contigs and singlets by sequence identity; transcripts showing 99.5% identity were classified as homoelogous genes, whereas those having 95% identity were grouped into orthologous gene clusters (162,874). The latter were used for BLAST searches against Poaceae transcripts, 50.2% of which were also found in barley, Brachypodium, or rice databases. We accordingly predicted 27,943 gene clusters in wheat that are commonly found in Poaceae. The remaining gene sets were searched against wheat genomic sequences. We obtained 13,060 gene hits, which were considered as wheat-specific genes. Of 284,822 genes, ca. 9,000 cDNAs were unconnected. Among these, we selected 4,400 EST clones and performed re-sequencing to obtain complete sequences of the inserts. This sequence information including full-length cDNA sequences is a useful genome resource for gene annotation of common wheat. This research was supported in part by a grant from the Ministry of Agriculture, Forestry and Fisheries of Japan (Genomics for Agricultural Innovation, KGS1002).