[BioC] DiffBind for Galaxy

[Rory Stark]

Hi Christoph-

Are the aligned read files you are using actually in BED format (chr,
start, end, name, score, strand)?

I'll look into adding an option to allow a .dat extension as an
alternative to .bed; if it is easy (and Gord has time) I'll try to have it
in the 2.13 release.

Cheers-
Rory


On 17/09/2013 16:23, "Christoph Grunau" <christoph.grunau at univ-perp.fr>
wrote:

>Dear Rory,
>
>thank you very much for your DiffBind software. It s actually the only
>software that allows us to identify differentially modified regions in
>ChIPseq against histone isoforms. I use it a lot and I would like to
>share it with my colleagues via our local Galaxy instance. Unfortunately,
>there is one mayor obstacle. As you will know, galaxy files have the
>uniform .dat suffixe. However for DiffBind bamReads and bamControl file
>format is detected via suffixes that must be bed, bam or bed.gz
>
>dba.count
>
>Count reads in binding site intervals
>
>Description
>Counts reads in binding site intervals. Files must be one of bam, bed and
>gzip-compressed bed. File
>suffixes must be ".bam", ".bed", or ".bed.gz" respectively.
>
>I solved this issue by creating a copy of the .dat file with suffix .bed.
>However, this slows down the analysis and generates redundancy for files
>that are relatively large. Would it be possible to modify your code so
>that the file format of bamReads and bamControl can be transmitted as
>parameter similar to what is used for loading the peakset?
>I know I ask a lot but this modification would be tremendously helpful.
>
>Many thanks and very best regards
>
>Christoph Grunau
>
>
>--
>Christoph Grunau
>Prof. des Universités/Professor (HDR)
>Université de Perpignan Via Domitia
>UMR 5244 CNRS Ecologie et Evolution des Interactions (2EI)
>52, avenue Paul Alduy
>66860 PERPIGNAN Cedex
>France
>Tel 33 (0)4.68.66.21.80
>Fax 33 (0)4.68.66.22.81
>http://2ei.univ-perp.fr/
>http://methdb.univ-perp.fr/epievo/
>
>
>