Two Bigwig files per sample are generated representing each strand

[drimran87]

Description of feature

The current rna-seq pipeline, generates two bigwig files one each for each strand. it would be great if like chipseq pipeline, it generate only one bigwig files, would be helpful when comparing chipseq and rnaseq data on IGV.

[MatthiasZepper]

Conversely, the strand-specific coverage information is what most people need, so I think there are good reasons to keep it like it is.

Consider using the IGV functionality Overlay Data Tracks to combine your tracks when viewing, using WiggleTools to combine them upfront or creating coverage tracks afterwards from your BAM files. You can also directly view your BAM files in IGV.

[drpatelh]

Yep, I agree. Most users will want to know what the coverage looks like based on the strandedness of the gene itself. We could have an implementation in the pipeline that generates all 3 bigwig files e.g. 1 for each strand and 1 combined. Be useful to know what command we would run with WiggleTools to do this with the current files created by the pipeline and we could make it an opt-in parameter to run this step.

[tdsone]

@drimran87 in case you don't need the processing steps downstream of the read mapping you can set --outWigStrand Unstranded to collapse the wig files into one.

Reference: https://github.com/alexdobin/STAR/blob/master/doc/STARmanual.pdf