[BioC] QuasR special case alignment

[Michael Stadler]

Dear Ugo,

One way to accomplish would be to put your two reporter genes into a
fasta file, and use this as your "genome" (see ?qAlign or the vignette
for examples of genomes in fasta format).

However, I think it would be better to align against the full genome, as
this will avoid certain problems, such as reads being suboptimally
aligned to your reporter genes which would align perfectly to some other
genomic locus.

Assuming that the reporter genes are part of the genome, you could then
count alignments using qCount() with a GRanges query containing the
ranges of your reporter genes.

If the reporter genes are not part of the genome, you can provide your
reporter gene sequences (as a fasta file) to qAlign, and all reads not
aligning to the genome will be further aligned to them. The format of
the required auxiliary file is very simple and described in section 5.2
of the vignette.

You can open the QuasR vignette using:
library(QuasR)
vignette("QuasR-Overview")

Best wishes,
Michael

On 25.04.2013 16:52, Ugo Borello wrote:
> Good morning,
> 
> I am doing RNA-Seq analysis and I would like to perform an exploratory
> analysis to first verify that 2 reporter genes are expressed in my samples.
> 
> Could the qAlign function perform an alignment of my short reads with two
> reporter gene sequences, excluding the genome.
> It seems, from my first attempt, that the 'genome' parameter is essential;
> is this true?
> Anyway how do I have to feed those sequences to qAlign and how do I have to
> format my auxiliary file/files to see in the alignment statistics the number
> of mapped reads for each reporter gene sequence separately?
> 
> Thank you in advance for your help.
> 
> Ugo
> 
> P.S. Let me say that QuasR is a fantastic package!
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>