Question: How to merge replicates of mapped reads out of Tophat for pathview and gage

[Parharn]

Hi, I have three biological replicates for my experiment. I wonder how should I merge the accepted_hits.bam out of Tophat for doing Pathview/gage analysis for "RNA-seq Pathway and Gene-set analysis workflow"? Let's name the accepted_hits.bam replicates as 1.bam 2.bam and 3.bam for the sake of short writing.

I have a thread on biostars.org

https://www.biostars.org/p/105503/

[TiborNagy]

The documentation suggest you need to use a differential expression analysis solution: DESeq2, Cufflinks, Limma and after that use gage. These packages handle the replicates.

[bigmw]

You would need to index all of them and get .bam.bai file for each one of them. Then collect all .bam and .bam.bai files into the same directory (i.e. tophat_all as in the demo code). Then all samples will be scanned in by function summarizeOverlaps at step 1: Count the reads mapped to each gene.
If you read carefully, actually all details are covered in Section 3.1 Preparation: read mapping and package installation:

http://www.bioconductor.org/packages/devel/bioc/vignettes/gage/inst/doc/RNA-seqWorkflow.pdf

Hi, I have three biological replicates for my experiment. I wonder how should I merge the accepted_hits.bam out of Tophat for doing Pathview/gage analysis for "RNA-seq Pathway and Gene-set analysis workflow"? Let's name the accepted_hits.bam replicates as 1.bam 2.bam and 3.bam for the sake of short writing.

I have a thread on biostars.org
https://www.biostars.org/p/105503/

[bigmw]

That’s the joint workflow for the users’ convenience. But GAGE/Pathview does not really require a external differential expression analysis step, as shown in the native workflow. For details:

http://bioconductor.org/packages/release/bioc/vignettes/gage/inst/doc/RNA-seqWorkflow.pdf

The documentation suggest you need to use a differential expression analysis solution: DESeq2, Cufflinks, Limma and after that use gage. These packages handle the replicates.

[Parharn]

Yes, later I discovered it but thanks to saying anyway