BEDtools in Galaxy

[Assaf Gordon]

Hi Aaron,

I'd like to add couple of your tools in our Galaxy server.
Have you (or other list members) tried that already (i.e. wrote an XML for some of your programs) ?

-gordon

[Aaron Quinlan]

Hey Gordon,

I have not done this yet, though we were just discussing it today.  We do not have a local Galaxy instance but are contemplating it...is it difficult to add "rogue" software?

Many of the BEDTools features are present in Galaxy, but the BAM functionality and BEDPE functionality are not. Do you intend to add support for these tools in order to build upon the existing tools?

[Assaf Gordon]

On 05/25/2010 06:59 PM, Aaron Quinlan wrote:
>
> I have not done this yet, though we were just discussing it today. We
> do not have a local Galaxy instance but are contemplating it...is it
> difficult to add "rogue" software?

Galaxy is very easy to extend, it's inherently built that way.

Here's how to install a local mirror:
http://bitbucket.org/galaxy/galaxy-central/wiki/GetGalaxy

Here's how to add a simple tool:
http://bitbucket.org/galaxy/galaxy-central/wiki/AddToolTutorial

>
> Many of the BEDTools features are present in Galaxy

But they are written in Python, and I feel they occasionally do not scale up with the size of the input datasets.
Also, in some cases your tools have added features beyond the galaxy tools.

> but the BAM functionality and BEDPE functionality are not.

BEDPE is a nice idea, but BAM I don't like that much. It's a personal preference of course, but I recommend to users of my local galaxy server to keep everything as textual files, not binary files. It allows greater flexibility within Galaxy (IMHO).

> Do you intend to add
> support for these tools in order to build upon the existing tools?

First priority for me is GenomeCoverageBed.
Then, intersect/merge/window/slop/closest BED, and coverageBed.
(Funny thing, the functionality of 'closestBed' is something that was requested several times on the galaxy mailing list).
Then, the paired-end stuff.

I can start by sending an XML for GenomeCoverageBed as a starting-point, and we'll continue from there.

-gordon

[Davide Cittaro]

Hi Aaron

On May 26, 2010, at 12:59 AM, Aaron Quinlan wrote:

Hey Gordon,

I have not done this yet, though we were just discussing it today.  We do not have a local Galaxy instance but are contemplating it...is it difficult to add "rogue" software?

Nope, it's pretty easy instead

Many of the BEDTools features are present in Galaxy, but the BAM functionality and BEDPE functionality are not. Do you intend to add support for these tools in order to build upon the existing tools?

Yes, they are. Nevertheless, I think BEDTools can perform much better. interval operations in galaxy rely on python scripts and bx-python module, the latter is pretty fast but slower than C++ software. I think one can add a BEDTools section with all the bedtools written so far.

d

On Tue, May 25, 2010 at 6:21 PM, Assaf Gordon <gor...@cshl.edu> wrote:

Hi Aaron,
I'd like to add couple of your tools in our Galaxy server.
Have you (or other list members) tried that already (i.e. wrote an XML for some of your programs) ?

-gordon

/*

Davide Cittaro

Cogentech - Consortium for Genomic Technologies

via adamello, 16

20139 Milano

Italy

tel.: +39(02)574303007

e-mail: davide....@ifom-ieo-campus.it

*/