RNA-sequencing

Overview

RNA-seq (RNA-sequencing) is a technique that helps to quantify RNA in a sample using next generation sequencing (NGS). It analyzes the transcriptome or gene expression pattern within the cell population. Here, we look at why RNA-seq is useful, how the technique works, and the basic protocols which are commonly used today. A typical RNA-seq experimental workflow involves the isolation of RNA from samples of interest, generation of sequencing libraries, use of a high-throughput sequencer to produce hundreds of millions of short paired-end reads, alignment of reads against a reference genome or transcriptome, and downstream analysis for expression estimation, differential expression, transcript isoform discovery, and other applications. There are several variations of the experimental protocol:

  1. mRNA-Seq vs. whole transcriptome: Researchers are usually interested in expression of mRNAs, but up to 95% of cellular RNA is rRNA. Thus, in order to not to waste reads on rRNA, researchers use one of two techniques: For mRNA-Seq, polyA+ RNAs are selected using oligo-dT beads. This excludes rRNAs, most of smRNAs and some mRNAs that are not polyadenylated (e.g. some of histone mRNAs). Also nascent mRNAs that have not been fully processed and polyadenylated will be dropped. Alternatively, for whole transcriptome sequencing, one can remove rRNAs by hybridization to rRNA-specific LNA probes with kits like Ribo-minus/Ribo-zero. This is less efficient, but keeps other RNAs, including unprocessed RNAs in the mixture. One can expect to see some of non-mRNA and mRNAs that have not been fully processed (e.g., with introns) in the results.
  2. Strand-specific (aka dUTP) vs. non-strand-specific. In order to be able to distinguish antisense expression, researchers use dUTP method to determine the strand from which the signal comes in RNA-Seq. Our pipeline is optimized the most-accepted approach, dUTP method. With this method the reads will map to the strand opposite to the RNA. Don't use this pipeline for miRNA-like strand-specific library cloning approach.

A typical gene expression and differential expression analysis has several common themes across different tool sets and RNA-seq analysis goals. RNA-seq analysis typically relies on inputs such as reference genome sequences, gene annotations, and raw sequence data. Working with these inputs requires familiarity with several standardized file formats such as FASTA (.fa), FASTQ, and gene transfer format (GTF). Typical RNA-seq analysis workflows start with raw data quality control (QC), then perform read trimming, alignment or assembly of reads, apply customized algorithms for a particular analysis goal (e.g., Cufflinks and Cuffdiff for gene expression analysis), and end with summarization and visualization of the results.

SciDAP RNA-Seq workflow

Current workflow consist of the following steps:

  1. Trim adapters from input FASTQ files.
  2. Use STAR to align reads from input FASTQ files according to the predefined reference indices; generate unsorted BAM file and alignment statistics file.
  3. Use fastx_quality_stats to analyze input FASTQ files and generate quality statistics files
  4. Use samtools sort to generate coordinate sorted BAM(+BAI) file pair from the unsorted BAM file obtained on the step 1 (after running STAR).
  5. Generate BigWig file based on the sorted BAM file. For the dUTP/strand-specific pipelines bigwig files for each strand will be produced. Please note that the reads map to the opposite strand.
  6. Map input FASTQ files to predefined rRNA reference indices using Bowtie to define the level of rRNA contamination; export resulting statistics to file.
  7. Calculate isoform expression level for the sorted BAM file and GTF/TAB annotation file using GEEP reads-counting utility; export results to a file.

Video performing RNA-Seq analysis in SciDAP

RNA-Seq

View RNA-Seq results in SciDAP

RNA-Seq