Thanks Jen! That makes more sense to me.

Can you explain more about the 'custom reference genome'?

Do you just mean uploading the reference genome in the normal fashion, (FTP or web form) and then it's available to the NGS tools?

-Greg

On Wed, Dec 19, 2012 at 12:46 PM, Jennifer Jackson jen@bx.psu.edu wrote:

Hi Greg,

At the very top of this wiki: http://wiki.galaxyproject.org/Admin/NGS%20Local%20Setup

is a link to a related wiki:

http://wiki.galaxyproject.org/Admin/Data%20Integration

that has additional required set-up info. This page also includes information about how to rsync the indexes currently available on the public Main server. These are planned to be updated with many new genomes and indexes over the next few weeks, so check back in mid January if you do not find what you want now and do not want to build it yourself. Releasing Bowtie2 indexes (also used by Tophat2) into this area is planned to start in later Jan, pending any issues, but to build yourself, the basic guidelines for Bowtie can be followed with adjustments to use " bowtie2-build" and " bowtie2_indices.loc". The Bowtie2 source tool documentation has the exact options and should be used along with our guidelines (this is true for all tools that require indexes).

If you choose not to populate genomes as an admin, users can use most tools with a 'custom reference genome'. Please be aware that running tools this way will take more processing resources and storage disk for user's histories. The size of the genome is the primary factor to consider - some mix of built-in and custom genomes may be good solution. Also, FTP upload will almost certainly need to be enabled. Complete "how-to" details from a user's perspective are here: http://wiki.galaxyproject.org/Support#Custom_reference_genome

Hopefully this gives you an option that will work,

Jen Galaxy team

On 12/18/12 9:40 AM, greg wrote:

Hi all,

I'm finishing up installing Galaxy on our local cluster.

I'm at this step regarding setting up reference genomes: http://wiki.galaxyproject.org/Admin/NGS%20Local%20Setup#Setting_Up_the_Refer...

But it looks like a whole lot of manual work that needs to be done by a system administrator. I was kind of expecting this would be something end users would be able to do. I'm actually not sure we have the resources for a sys admin to install all of these reference genomes every time someone needs one.

I guess here are my questions:

Am I misunderstanding the process?

Is there a different way that normal users can add reference genomes instead?

Is there anything in the works that lets users easily add these?

Is there a way to use NGS tools without installing reference genomes?

Thanks,

Greg ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at:

http://lists.bx.psu.edu/

-- Jennifer Jackson http://galaxyproject.org