I linked the macs version 1.3.7.1 to the default as the commands below.

It runs, however, still get the error message as below --

MACS on data 65 and data 66 (peaks: bed)

0 bytes

An error occurred running this job: INFO @ Fri, 11 Jan 2013 19:30:50:

# ARGUMENTS LIST:

# name = MACS_in_Galaxy

# format = SAM

# ChIP-seq file = /mnt/galaxyData/files/000/dataset_66.dat

# control file = /mnt/galaxyData/files/000/dataset_65.dat

# effective genome size = 2.70e+09

# tag size

I switched to version 1.4.1 and got the error message same as the first time.

Thanks again,

Kathryn

-----Original Message-----
From: Dannon Baker [mailto:dannonbaker@me.com]
Sent: Friday, January 11, 2013 12:12 PM
To: Jennifer Jackson
Cc: Galaxy Dev; Sun, Wenping [USA]
Subject: [External] Re: macs error for chipseq data

There's an issue with the MACS installation on the current cloud tools volume (which will be fixed with the next volume update coming soon).

For existing instances, you can get MACS working correctly by executing the following two commands (which change the default version of macs used) after ssh'ing in to your instance:

sudo su galaxy

ln -sfn /mnt/galaxyTools/tools/macs/1.3.7.1 /mnt/galaxyTools/tools/macs/default

And that's it, no need to restart galaxy or anything.

-Dannon

On Jan 11, 2013, at 10:59 AM, Jennifer Jackson <jen@bx.psu.edu> wrote:

> Hi Kathryn,

> I am going to post this back to the galaxy-dev list and cc Dannon so that he or or one of the developers more experienced with troubleshooting cloud issues can help. My guess is that there is a path problem (covered in the first wiki link) but by no means is that the only possibility.

> Thanks!

> Jen

> Galaxy team

> On 1/11/13 7:38 AM, Sun, Wenping [USA] wrote:

>> Hi Jen,

>> Sorry that I missed some part of your email. It is not for local galaxy install. I am using the cloudman from aws and it seemed to have everything for having the input files and reference genome chose. The job was shown on the right panel and then the error message coming out.

>> It seems to me that MACS was installed…. What is the issue?

>> Thank you very much,

>> Kathryn

> On 1/11/13 6:55 AM, Sun, Wenping [USA] wrote:

>> Hi Jennifer,

>> Thanks for the information on MACS tools.

>> Yes, I forgot to mention that I used bowtie mapped files as the input.

>> Thank you very much,

>> Kathryn

>> From: Jennifer Jackson [mailto:jen@bx.psu.edu]

>> Sent: Friday, January 11, 2013 3:48 AM

>> To: Sun, Wenping [USA]

>> Cc: galaxy-user@bx.psu.edu

>> Subject: [External] Re: [galaxy-user] macs error for chipseq data

>> Hello Kathryn,

>> This is occurring on your local Galaxy install? Have you installed the actual MACS tool and set up the proper configuration paths? Details for how to do this are in these wikis:

>> http://wiki.galaxyproject.org/Admin/Config/Tool%20Dependencies

>> http://wiki.galaxyproject.org/Admin/Tools/Tool%20Dependencies

>> Apart from that, I should let you know that prior to running MACS, groomed fastq datasets first need to be mapped. An example can be found in our Using Galaxy paper, protocol #3. Links to the paper and supplemental materials (including a screencast walk-through) are available here:

>> https://main.g2.bx.psu.edu/u/galaxyproject/p/using-galaxy-2012

>> If you need more help with your local install, directing questions to

>> the galaxy-dev@bx.psu.edu mailing list would be best,

>> Thanks!

>> Jen

>> Galaxy team

>> On 1/10/13 1:14 PM, Sun, Wenping [USA] wrote:

>> Dear galaxy users,

>> I’ve encountered the error while running macs on galaxy for chipseq

>> data. I used fastq groomed files as input and had the following error

>> An error occurred running this job: /bin/sh: macs: not found

>> Anybody can kindly provide some hints?