How to convert strand specific RNASEQ bam to bigwig

Question

[a.bafna]

Hi, I am working with dUTP bases stranded RNA-Seq paired end data. After aligning with STAR , I would like to convert my bam files to bigwig. I am trying to use bamCoverage but cannot use the --filterRNAstrand in Galaxy. Please can anyone help . Thanks

[jv]

why are you unable to use --filterRNAstrand? Do you get an error when trying to use this parameter? If so, what is the error?

[a.bafna]

Hi , In usegalaxy.eu instance there is no option for --filterRNAstrand. Hence I cannot filter them.

[jv]

Sounds like deeptools on that galaxy instance is out of date.

All --filterRNAStrand really does is selects reads that have mapped to one strand and returns a bigwig file with coverage of those reads. (–filterRNAstrand=forward keeps minus-strand reads, which originally came from genes on the forward strand using a dUTP-based method)

You can do this yourself using samtools view to select the minus-strand reads from your BAM file and use that as input to bamCoverage to generate your plus-strand coverage bigwig file, then repeat with plus-strand reads.

[Carlo Yague]

You can do this yourself using samtools view to select the minus-strand reads from your BAM file and use that as input to bamCoverage to generate your plus-strand coverage bigwig file, then repeat with plus-strand reads.

Its not that simple with paired-end reads (because usually, the mates are on different strands), see How To Separate Illumina Based Strand Specific Rna-Seq Alignments By Strand

Answer 1

[a.bafna]

Hi, thanks for reply. I indeed used "Only include reads originating from fragments from the forward or reverse strand" option and selected reverse . Now I assumed it will only count the plus-strand reads, which originally came from genes on the reverse strand using a dUTP-based method) On viewing the result in UCSC Browser I get the coverage as http://genome.ucsc.edu/cgi-bin/hgTracks?db=mm10&lastVirtModeType=default&lastVirtModeExtraState=&virtModeType=default&virtMode=0&nonVirtPosition=&position=chr1%3A91405146%2D91470160&hgsid=1355475517_CvVtrd8za2fLHooDlDFACr2ZAp72. Is that what I should select and expect. I was expecting the reads to be in downward direction as given after wigtobigwig conversion for reverse reads.