Visualize Strand-Specific Rna-Seq : How To Differentiate Strands

Question

[Nicolas Rosewick]

Hi,

I wonder how to display such a figure ( the green coverage). So how differntiate the sens coverage from the anti-sens coverage ? I search in IGV options but I didn't find anything ...

Thanks a lot

N.

[Istvan Albert]

One possible workaround would be to split the bam file by strand and load each separately.

Answer 1

[Ann]

IGB bioviz.org) can display different strands data in different tracks. To separate plus and minus strand alignments into two tracks, select the track, select Annotation tab, and deselect the combine strand option ( +/- ) in the section labeled "Strand." It can also color-code the data by strand. Here's an image to illustrate:

https://www.dropbox.com/s/sp7zqat7r2n568u/SeparateStrands.png

[Antonio R. Franco]

this is why I like IGB..

Answer 2

[Steve Lianoglou]

As far as IGV is concerned, I believe that it does not have this feature (yet).

If you want to use other tools, the bioconductor package Gviz has an AlignedReadTrack which plots stranded coverage data in a similar fashion to the green chunk of the image you posted above.

Answer 3

[Geparada]

One simple way to do this on any visualization program it's just pre-filter the SAM/BAM by strand, and put the reads that aling to minus or plus strand in two different files. Then, you can visualize the coverage of both strand by separate.

[Nicolas Rosewick]

I tried with samtools but it seems not to be ok :

for positive strand

samtools view -F 0x10 -b input.bam > postiveStrand.sam

for negative strand

samtools view -f 0x10 -b input.bam > negativeStrand.sam

[Istvan Albert]

make sure to use the right flag, and you can just use integers 16 in this case see http://picard.sourceforge.net/explain-flags.html

[Nicolas Rosewick]

What I want to do is to extract the reads (and its pair) where the first-in-pair (so coming from the R1 fastq file) aligned to the reverse strand . Is that possible ?

[Carlo Yague]

See this cool tutorial by Istvan Albert.

Answer 4

[wm]

@Ann How can I show the strand-specific reads? This fig seems not a strand-specific alignment (both strand have almost the same reads).

Answer 5

[jerry]

If you have paired-end reads, then just splitting the BAM file into forward and reverse strand BAM files using samtools view will NOT work. All this will do is place all forward 'mate' reads on the forward strand, and reverse mate reads on the reverse strand, which is not what you want. You want to know which paired-end ALIGNMENTS (not reads!) map to the forward strand, and which alignments map to the reverse strand. Unfortunately, this information is hard to extract from the BAM file, and IGV does not have this feature yet.

[Chirag Nepal]

so how to do it? any idea.

[erwan.scaon]

Very interested here too ! Did someone make any progress on this question ?