Abstract
In the genome projects, protein coding genes are computationally predicted and annotated using many tools. On the other hand, annotation information of non-coding RNA (ncRNA) is not found in the most of published genome sequences. The reason is we don't have tools for gene prediction and analysis of ncRNA. tRNA is one of the ncRNAs and has significant function on translation. Their copy number is several hundreds and transcribed by Pol III. Since tRNA transcripts form rigid secondary (clover-leaf) structure and have highly conserved promoter motifs that we can easily find location of genes from genome sequences. However, we don't predict which tRNA gene is functional or pseudogene because of lacking gene characterization with experimental analysis. Therefore we have made genome-wide transcription assay from rice and Arabidopsis genes for increasing of information to build tRNA gene prediction algorithm. Furthermore, we have made comparison between rice and Arabidopsis tRNA gene organization.
