The identification and characterization of novel transcripts from RNA-seq data

Abstract

Owing greatly to the advancement of next-generation sequencing (NGS), the amount of NGS data is increasing rapidly. Although there are many NGS applications, one of the most commonly used techniques 'RNA sequencing (RNA-seq)' is rapidly replacing microarray-based techniques in laboratories around the world. As more and more of such techniques are standardized, allowing technicians to perform these experiments with minimal hands-on time and reduced experimental/operator-dependent biases, the bottleneck of such techniques is clearly visible; that is, data analysis. Further complicating the matter, increasing evidence suggests most of the genome is transcribed into RNA; however, the majority of these RNAs are not translated into proteins. These RNAs that do not become proteins are called 'noncoding RNAs (ncRNAs)'. Although some time has passed since the discovery of ncRNAs, their annotations remain poor, making analysis of RNA-seq data challenging. Here, we examine the current limitations of RNA-seq analysis using case studies focused on the detection of novel transcripts and examination of their characteristics. Finally, we validate the presence of novel transcripts using biological experiments, showing novel transcripts can be accurately identified when a series of filters is applied. In conclusion, novel transcripts that are identified from RNA-seq must be examined carefully before proceeding to biological experiments.

Epigenomic and transcriptomic approaches in the post-genomic era: path to novel targets for diagnosis and therapy of the ischaemic heart? Position Paper of the European Society of Cardiology Working Group on Cellular Biology of the Heart

“…strand specific sequencing), as well as the depth of the sequencing and bioinformatics analysis can yield strikingly different results depending on, e.g. the mapping to the reference genome and annotations 100 . Such technical details may become major obstacles when comparing independent studies, which not only use different experimental conditions and platforms, but also varying bioinformatics approaches making potential meta-analyses difficult or useless.…”

Sexual stage-induced long noncoding RNAs in the filamentous fungus Fusarium graminearum

“…To discover lncRNAs expressed during the perithecia development, we adopted an established protocol for novel transcripts identification, with some modifications (Weirick et al 2016;Supplemental Fig. S4).…”

High-Throughput Methods to Detect Long Non-Coding RNAs

“…It is absolutely true that unknown (and novel) lncRNAs will be discovered via RNA-seq than microarrays. However, a caution is needed as many of such novel lncRNAs contain many repetitive elements as we recently reported [40]. Furthermore, the mapping rate of sequence reads depends on the version of reference genome, such that the most recent human genome assembly GRCh38/hg38 has much better mapping rate than GRCh37/hg19 [40].…”

“…However, a caution is needed as many of such novel lncRNAs contain many repetitive elements as we recently reported [40]. Furthermore, the mapping rate of sequence reads depends on the version of reference genome, such that the most recent human genome assembly GRCh38/hg38 has much better mapping rate than GRCh37/hg19 [40]. In addition, the computational power needed to analyze RNA-seq data is much more demanding than that of microarrays.…”