Data Analysis Pipeline for RNA‐seq Experiments: From Differential Expression to Cryptic Splicing

Abstract

RNA sequencing (RNA-seq) is a high-throughput technology that provides unique insights into the transcriptome. It has a wide variety of applications in quantifying genes/isoforms and in detecting non-coding RNA, alternative splicing, and splice junctions. It is extremely important to comprehend the entire transcriptome for a thorough understanding of the cellular system. Several RNA-seq analysis pipelines have been proposed to date. However, no single analysis pipeline can capture dynamics of the entire transcriptome. Here, we compile and present a robust and commonly used analytical pipeline covering the entire spectrum of transcriptome analysis, including quality checks, alignment of reads, differential gene/transcript expression analysis, discovery of cryptic splicing events, and visualization. Challenges, critical parameters, and possible downstream functional analysis pipelines associated with each step are highlighted and discussed. This unit provides a comprehensive understanding of state-of-the-art RNA-seq analysis pipeline and a greater understanding of the transcriptome. © 2017 by John Wiley & Sons, Inc.

Tau-mediated Disruption of the Spliceosome Triggers Cryptic RNA-splicing and Neurodegeneration in Alzheimer’s Disease

“…Gene level counts were obtained as the sum of total reads mapped to respective genes. As expected, based on principal component analysis using gene counts (Yalamanchili et al, 2017)., sample clustering reflected genotype and age ( Figure S7).…”

Forniceal deep brain stimulation induces gene expression and splicing changes that promote neurogenesis and plasticity

“…DEIs were computed using the pipeline described in ( Yalamanchili et al, 2017 ). Isoform expression was quantified from raw pair-end fastq files (RNA-Seq data) using Kallisto ( Bray et al, 2016 ), an alignment free transcript quantification program.…”

Partial loss of CFIm25 causes aberrant alternative polyadenylation and learning deficits

“…We sequenced four samples from both the control and NUDT21 knockdown conditions, but excluded one control sample with <10 million reads, and one knockdown sample with unusually low NUDT21 levels (<2% of other knockdown samples). We then performed principle component analysis (PCA) on gene counts as previously described and confirmed that the treatment groups clustered ( Figure 4—figure supplement 1A ) (Yalamanchili et al, 2017).…”

“…We performed a PCA as previously described to confirm that the treatment groups separated ( Figure 4—figure supplement 1B ) (Yalamanchili et al, 2017). We then computed protein level changes as log 2 fold change values and plotted them against their relative mRNA length change ( Figure 4E ).…”