I agree, I prefer your solution, I will focus on that solution, thanks!

Although there is some software more or less used in the community such Delly https://github.com/tobiasrausch/delly and Breakdancer http://gmt.genome.wustl.edu/packages/breakdancer/documentation.html, that doesn't use bed files, the only way to parallelize their execution is through smaller bams

Regards

On 6 May 2015 at 15:00, Peter Cock <p.j.a.cock@googlemail.com> wrote:

On Wed, May 6, 2015 at 11:33 AM, Roberto Alonso CIPF <ralonso@cipf.es> wrote:
> Hello,
>
> I agree, what you say fits perfectly for GATK, but as I wanted to create a
> more generic code I did it this way (also because I am a newbie in the
> galaxy code and I didn't know so well how to implement this ). What about a
> tool that doesn't accept a region, just a bam? Maybe we can put another
> parameter in the parallelism tag that force to split the bam.
> Mostly, just to create a bed file would be better, right?
> What do you think?
>
> Regards

Maybe you're right - BAM splitting might be useful for some tools
(any examples?), even though BED splitting is a much more elegant
solution.

Peter

Roberto Alonso

Functional Genomics Unit
Bioinformatics and Genomics Department
Prince Felipe Research Center (CIPF)

C./Eduardo Primo Yúfera (Científic), nº 3
(junto Oceanografico)
46012 Valencia, Spain
Tel: +34 963289680 Ext. 1021
Fax: +34 963289574
E-Mail: ralonso@cipf.es