Assessment of transcript reconstruction methods for RNA-seq

By Tamara Steijger, J. F. Abril, Pär G. Engström, F. Kokocinski, Tim J. P. Hubbard, Roderic Guigó, J. Harrow, Paul Bertone. Nature Methods, 2013

684 citations recorded by the source corpus.

We evaluated 25 protocol variants of 14 independent computational methods for exon identification, transcript reconstruction and expression-level quantification from RNA-seq data. Our results show that most algorithms are able to identify discrete transcript components with high success rates but that assembly of complete isoform structures poses a major challenge even when all constituent elements are identified. Expression-level estimates also varied widely across methods, even when based on similar transcript models. Consequently, the complexity of higher eukaryotic genomes imposes severe limitations on transcript recall and splice product discrimination that are likely to remain limiting factors for the analysis of current-generation RNA-seq data.

Subfields: Cancer Research, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/a16c7227dc7a36932ddde034054dd6779d63980c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cancer Research

In Molecular Biology

Citations in the corpus

StringTie enables improved reconstruction of a transcriptome from RNA-seq reads (2015, background): “A recent study 18 of the current transcript reconstruction methods showed that even in cases where these methods identified all constituent exons of a transcript, they often failed to assemble the exons into complete isoforms.”
Integration of mapped RNA-Seq reads into automatic training of eukaryotic gene finding algorithm (2014, methodology): “The RNA-seq Genome Annotation Assessment Project (RGASP) Consortium (8) comprehensively assessed the accuracy of transcript reconstruction from RNA-Seq data by several algorithms.”
Data quality aware analysis of differential expression in RNA-seq with NOISeq R/Bioc package (2015, background): “05 TMM (edgeR, NOISeq), Quantiles, Upper Quartile (EDASeq, NOISeq), DESeq, DESeq2 4 Filtering out low count features It has been often argued that, in RNA-seq, expression estimation for low count genes is less reliable because read counts could have been assigned by chance [1, 2].”
Developmental regulation of human cortex transcription and its clinical relevance at base resolution (2014, neutral): “Such feature-based read counts lack the ability to reliably identify novel transcriptional activity, but generally limit the inherent difficulty in transcript assembly and characterization based on short read sequencing technologies 8 .”
Current challenges and best-practice protocols for microbiome analysis (2019, background): “Apparently, it also helps in sequencing entire RNA transcripts and provides precise information on the existence of specific isoforms [138].”
Gaining comprehensive biological insight into the transcriptome by performing a broad-spectrum RNA-seq analysis (2017, background): “We observed that, unlike previous studies 2 , in more challenging examples like MCF7-300, STAR reported a much higher number of transcripts (mostly single exons) but with a high FP rate (Fig.”
BRAKER1: Unsupervised RNA-Seq-Based Genome Annotation with GeneMark-ET and AUGUSTUS (2016, background): “However, the RNA-Seq Genome Annotation Assessment Project (RGASP) (Steijger et al., 2013) has shown that transcriptome assembly is prone to errors.”
Predicting Genes in Single Genomes with AUGUSTUS (2018, methodology): “…for demonstration purposes you can use the file rgasp_fly_2R.fastq which contains RNA-seq reads of D. melanogaster from the RGASP assessment (Steijger et al., 2013) that map to the above mentioned genome file (when mapped with tophat2): bash input wget…”
NONCODE 2016: an informative and valuable data source of long non-coding RNAs (2015, methodology): “The complexity of higher eukaryotic genomes imposes severe limitations on transcript recall and splice product discrimination that are likely to remain limiting factors for the analysis of current-generation RNA-seq data (42).”
PacBio Sequencing and Its Applications (2015, methodology): “Personal transcriptomes are expected to have applications in understanding individual biology and disease, but SGS has been shown to be insufficiently accurate for the identification and quantification of an individual’s genetic variants and gene isoforms [53,63,64].”
Identification of protein coding regions in RNA transcripts (2014, methodology): “A comprehensive assessment of the accuracy of transcript reconstruction from RNA-Seq reads was conducted in the RGASP competition (3).”
SQANTI: extensive characterization of long-read transcript sequences for quality control in full-length transcriptome identification and quantification (2017, background): “However, both technologies have serious limitations for the reconstruction of the actual expressed transcripts, as short reads break the continuity of the transcript sequences and fail to resolve assembly ambiguities at complex loci 21,22 .”