Visualizing isoforms in RNA-seq data

One of the advantages of RNA-seq data compared to microarrays is that you get isoform-level information ‘for free’ in addition to gene-level information. In this exercise, we are going to look at some ways to identify and visualize isoforms.

As you have hopefully read on the introduction page, we will use RNA-seq and quantitative mass-spectrometry (MS) data sets from the A431 cell line. These data were measured by a group at SciLifeLab Stockholm in a ‘proteogenomics’ study where the aim was to discover new genes or gene variants by deep proteomic profiling using an MS method, and mapping the obtained peptides back to the genome. The RNA-seq data was obtained to see if there was RNA-level support for the predicted novel peptides and transcript variants. We will look at one locus that was flagged by the research group as being interesting, and see what the RNA-seq data look like for that gene.

Strategies for using the RNA-seq data

There are different ways to find out how the RNA-seq data shows the RAB11FIP5 gene to be expressed. Roughly speaking, we can talk about three different strategies:

• Mapping the reads to a reference genome and quantifying the gene and isoform FPKMs
• Mapping the reads to a reference genome and assembling transcripts based on the mapping (reference guided assembly)
• Assembling the reads de novo, ignoring the reference genome for now

In order to make these steps computationally feasible during the lab, we have extracted only those sequences that mapped to the RAB11FIP5 gene in each sample. These “sub-FASTQ” files can be found in /proj/b2013006/webexport/downloads/courses/RNAseqWorkshop/isoform/RAB11FIP5_fastqFiles.

To do the reference guided assembly yourself go to Reference guided assembly using Cufflinks or Stringtie.

This link also contains information on how to quantify already annotated genes and isoforms.

To do the de novo assembly yourself go to Isoform detection using RNA-seq de novo Assembly.

Visualise isoform data

For the gene RAB11FIP5 you will now hopefully have generated your own data that you can look at. If everything worked you will now have:

• One or two BAM files with short reads mapped to the genome using HISAT2 and/or STAR

• One GTF file containing different isoforms of RAB11FIP5 assembled with Cufflinks and/or Stringtie based on the short reads that mapped to the genome

• One BAM file with Trinity assembled transcripts mapped to the genome

Importing reference based isoform info for the RAB11FIP5 gene

Since it takes time to generate all data, we have already created other files that you can also download and view in your browser. This includes result files for the subset of reads that map to the RAB11FIP5 gene. These mappings have been used for reference based assembly of isoforms.

You can find all BAM files and GTF files for all samples here using a web browser. You can then view them in IGV using “Import from URL”.

Importing de novo assembled transcripts mapped to the RAB11FIP5 gene

We have also created result files from de novo transcriptome assembly using the the subset of reads that map to the RAB11FIP5 gene. The assembled transcripts were then mapped back to the genome.

You can download all BAM files and GTF files for all samples here using a web browser.

Importing reference based isoform info to the genome

In addition, we provide result files for all reads that were mapped to the genome. These mappings were used for reference based assembly of isoforms across the entire genome. There is a GTF file with the final merged isoform
information from all 12 samples.

You can download all BAM files and GTF files for all samples here using a web browser.

Importing the peptide track for the RAB11FIP5 gene and the genome

As mentioned above, we will look at some identified peptides from a mass-spectrometry experiment, and compare those with RNA-seq data from the same cell line.

You can download the BED file containing all peptides mapped to the genome here.

From the name of the BED file (human_A431_global-TDA-FDR1pc_green-known_red-novel.bed), IGV will automatically know to color the track according to peptide status (green for annotated peptides, red for novel peptides).

Importing the Pac bio reads mapped to the genome

You can use a web browser to access the BAM file containing all PacBio reads mapped to the genome here. Unfortunately there are no PacBio reads that mapped to the RAB11FIP5 gene, but you can look at other loci.

How do the PacBio reads from full length transcripts compare to the transcripts assembled from short reads?