High-quality reference transcriptome construction improves RNA-seq quantification in indica.

The Reference Transcriptomic Dataset (RTD) is an accurate and comprehensive collection of transcripts originating from a given organism. It holds the key to precise transcript quantification and downstream analysis of differential expressions and regulations. Currently, transcriptome annotations for most crop plants are far from complete. For example, indica ( indica) is reported to have 40,759 transcripts in the database without alternative transcript isoforms and alternative splicing (AS) events. To generate a high-quality RTD, we conducted RNA sequencing of rice leaf samples collected at various time points during infection. The obtained reads were analyzed by adopting the recently developed computational analysis pipeline to assemble the RTD with increased transcript and AS diversity for indica (IndicaRTD). After stringent quality filtering, the newly constructed transcriptome annotation was comprised of 122,968 non-redundant transcripts from 53,695 genes. This study identified many novel transcripts compared to deposited data that are important for regulating molecular and physiological processes in the plant system. Currently, the assembled IndicaRTD must allow fast quantification of transcript and gene expression with high precision.