P0051 Identification of Rhizome–Specific Transcripts, Putative TFs and SNPs in Ginger using Next-Generation Sequencing Technology

Min-Jeong Kim , Institute of Biological Chemistry, Washington State University, Pullman, WA
Ruifeng He , Institute of Biological Chemistry, Washington State University, Pullman, WA
David R. Gang , Institute of Biological Chemistry, Washington State University, Pullman, WA
Rhizomes are underground stems that serve several purposes, from vegetative propagation, to new territory invasion, to bioactive compound storage. Ginger is well known for the medicinal and culinary uses of its rhizomes. Recent advances in next-generation sequencing technology now greatly enable genomic scale investigations of non-model organisms such as ginger. To further our efforts to identify genes involved in rhizome growth, development, differentiation, and metabolism, we applied RNA-seq and total transcriptome analysis to rhizome and other ginger tissues. In order to survey rhizome-specific transcripts, we constructed ten ginger Illumina sequencing libraries from leaf and root tissues for single-read RNA-seq analysis and compared these to two total transcriptome paired-end read libraries from rhizome apical tip and rhizome elongation zone tissues. Assembly of sequencing reads produced 67,996, 83,682 and 73,123 unitrans (unique transcripts) from leaf, root and mixed rhizome libraries, respectively, from which we predicted 49,872, 62,352 and 62,866 protein coding ORFs using ESTScan. 426 unitrans were exclusively composed of rhizome reads. The most abundant GO slim groups were chloroplast in cellular component, nucleotide binding in molecular function, and protein metabolism in biological process categories. We also found unitrans representing putative TFs from each tissue: 1,764 from leaf, 2,030 from root and 1,736 from rhizome. The most abundant TFs were MYB in leaf and root, but bHLH in rhizome. Interestingly, the LFY was found only in rhizome. Additionally, we identified 22,145 rhizome-specific SNPs supported by at least 100 sequencing reads. As expected, the transition mutations (65.48%) were more abundant than transversions (34.52%).