A high-resolution single-molecule sequencing-based Arabidopsis transcriptome using novel methods of Iso-seq analysis

Abstract

Background Accurate and comprehensive annotation of transcript sequences is essential for transcript quantification and differential gene and transcript expression analysis. Single-molecule long-read sequencing technologies provide improved integrity of transcript structures including alternative splicing, and transcription start and polyadenylation sites. However, accuracy is significantly affected by sequencing errors, mRNA degradation, or incomplete cDNA synthesis. Results We present a new and comprehensive Arabidopsis thaliana Reference Transcript Dataset 3 (AtRTD3). AtRTD3 contains over 169,000 transcripts—twice that of the best current Arabidopsis transcriptome and including over 1500 novel genes. Seventy-eight percent of transcripts are from Iso-seq with accurately defined splice junctions and transcription start and end sites. We develop novel methods to determine splice junctions and transcription start and end sites accurately. Mismatch profiles around splice junctions provide a powerful feature to distinguish correct splice junctions and remove false splice junctions. Stratified approaches identify high-confidence transcription start and end sites and remove fragmentary transcripts due to degradation. AtRTD3 is a major improvement over existing transcriptomes as demonstrated by analysis of an Arabidopsis cold response RNA-seq time-series. AtRTD3 provides higher resolution of transcript expression profiling and identifies cold-induced differential transcription start and polyadenylation site usage. Conclusions AtRTD3 is the most comprehensive Arabidopsis transcriptome currently. It improves the precision of differential gene and transcript expression, differential alternative splicing, and transcription start/end site usage analysis from RNA-seq data. The novel methods for identifying accurate splice junctions and transcription start/end sites are widely applicable and will improve single-molecule sequencing analysis from any species.

BaRTv2: a highly resolved barley reference transcriptome for accurate transcript‐specific RNA‐seq quantification

“…This overcomes mis-annotation of ORFs and distinguishes protein-coding and unproductive isoforms (Brown et al, 2015;Zhang et al, 2017). The level of unproductive transcript isoforms of protein-coding genes in BaRTv2.18 (35.2%) compared well to the level in the recent Arabidopsis AtRTD3 (41%; Zhang et al, 2022). The translations from TranSuite also allow accurate identification of positions of PTCs and other NMD signals in unproductive transcripts.…”

“…Stringently filtered, and introduce errors (over-correction). We have used the novel Iso-seq analysis pipeline based on that carried out for Arabidopsis AtRTD3 (Zhang et al, 2022) with some modifications for the construction of BaRTv2.18. We used TAMA for greater control of analysis, a SJ centric approach to identify and remove false SJs, and a probabilistic approach to determine TSS and TES, which take transcript abundance into account.…”

“…Mapping FLNCs to the genome can be inaccurate due to the phenomenon of edge wander (Holmes & Durbin, 1998) where inaccuracies in defining the boundaries of a gap are the main source of error, for example when aligning a sequence with spliced intron to a genomic reference. Mis-mapping of reads containing spliced introns is exacerbated by local sequence errors (AE 10 nt) that frequently generate false SJs leading to transcripts with incorrect intron-exon boundaries (Zhang et al, 2022). To address this issue, a HC SJ dataset was created based on both Iso-seq and short-read RNA-seq data (see below).…”

“…For many genes, there were long reads representing full-length transcripts, but also 5 0 and 3 0 truncated reads likely due to degradation or incomplete cDNA synthesis. To distinguish between real and false transcript end sites, we used two methods depending on expression level (Zhang et al, 2022). We assumed that for genes with multiple reads with different ends, the start and ends of fragmentary reads would be distributed randomly and uniformly between the 5 0 -and 3 0 -ends, while true TSS/TES would have more read support than what would be expected by random.…”

“…Major challenges remain in the accurate definition of SJs and TSS and TES due to sequencing errors and degradation of mRNAs in vivo or during processing. We applied newly developed methods of Iso-seq analysis, which use sets of HC SJs and TSS/TES to define these key features (Kuo et al, 2020;Zhang et al, 2022). Some FLNC reads still have relatively high error rates (about 3%; Kuo et al, 2020).…”