Question: DiffBind on Galaxy
1
gravatar for j.m.fustin
13 months ago by
j.m.fustin10
j.m.fustin10 wrote:

Hi,

I am trying to perform Diff Bind on Galaxy using output from MACS and bam files for Read and Control. For the Read and control, I tried using raw tophat output bam files as well as the same bam files but sorted according to genomic coordinate. In both cases I get this message:

error

An error occurred with this dataset:

WARNING: ignoring environment value of R_HOME [1] "en_US.UTF-8" Fatal error: An undefined error occured, please check your intput carefully and contact your administrator. Loading required package: GenomicRanges Loading required package: methods Loading

Is it something I do wrongly?

Thanks in advance for your help!

 

JM

error en_us.utf-8 • 572 views
ADD COMMENTlink modified 6 days ago by zachary.gaber60 • written 13 months ago by j.m.fustin10
2
gravatar for Jennifer Hillman Jackson
13 months ago by
United States
Jennifer Hillman Jackson21k wrote:

Hello,

Where are you running the tool? If at Galaxy Main (http://usegalaxy.org) - or if this can be reproduced there - we can help to double check that the inputs are correct (or see below for items to confirm). Send in a bug report from the red error dataset and include a link to this post.

Other than that, or if on a local and want to troubleshoot yourself, I suggest double checking the tool install was complete (with included dependencies) and that your instance is up to date (v 16.01) first. Then examine the input contents and format. In particular, check that the reference genome for all inputs is a match (exact match - content in the files and plus assigned "database" attribute for datasets). From there, double check the peak input to make sure that the formatting is a match for the type specification. And finally, check that the data entered for sample, tissue, etc. all meet the specifications on the tool form.

I personally haven't seen this particular error before, but the above is how I would start to investigate the error on a local (for the tool/Galaxy build) and if that is intact, how I would confirm the inputs (above is what I would look at in a submitted bug report).

Thanks, Jen, Galaxy team

ADD COMMENTlink written 13 months ago by Jennifer Hillman Jackson21k

I found your bug report sent in earlier. Am reviewing.

ADD REPLYlink modified 13 months ago • written 13 months ago by Jennifer Hillman Jackson21k

Jen if it of any help please send me the error message.

ADD REPLYlink written 12 months ago by Bjoern Gruening4.5k

Hello, I would also like to say I have been experiencing the exact same problem with Diffbind and have been getting the same error message. I was wondering if you came up with a solution and or found out what the problem was. I've also submitted a bug report with a link to this thread. I've been running Diffbind with the peak file from MACS. I've tried a few different things for the Read and Control files - the SAM files that came out of Bowtie, the BAM files that I converted those SAM files into, and (what I really want to analyze), BAM files that had their PCR duplicates removed. In all cases I got the same error message. I'd apprecaite any help or suggestions you or anyone else has. Thanks.

ADD REPLYlink written 12 months ago by zachary.gaber60
0
gravatar for j.m.fustin
12 months ago by
j.m.fustin10
j.m.fustin10 wrote:

Hi Zachary, Hi Jen, I believe the problem is in the automatic creation of the csv file describing the samples and contrasts for running DiffBind. This csv file should be encoded in "en_US.UTF-8" (as you can set up on a MAC) but apparently it is not, either because of the galaxy system or because of our own computer during input (former is more likely). I'll try to run DiffBind@Galaxy from a MAC (I used PC previoulsy) and see if it makes a difference...let you know shortly.

ADD COMMENTlink written 12 months ago by j.m.fustin10
0
gravatar for j.m.fustin
12 months ago by
j.m.fustin10
j.m.fustin10 wrote:

...and no, it does not. It is thus probably within the Galaxy system.

ADD COMMENTlink written 12 months ago by j.m.fustin10
0
gravatar for christophe.habib
8 months ago by
France
christophe.habib330 wrote:

Hello,

I installed DiffBind (2016-03-08, revision d6dcb3dd5f8f) on my local galaxy instance (16.01) and I also have this error.

I think the relevant parts are :

"WARNING: ignoring environment value of R_HOME
[1] "en_US.UTF-8""

and

"Error in pv.DBA(DBA, method, bSubControl, bFullLibrarySize, bTagwise = bTagwise,  : 
  Unable to perform analysis: no contrasts specified."

But since I'm not really used to this tool I don't what are the contrast specified in this error.

Did you find a solution for the first problem ? Maybe it could solve the second one as well.

Thank you for your help.

ADD COMMENTlink modified 8 months ago • written 8 months ago by christophe.habib330

I forgot to say the most important part actually.
The dependency package_deseq2_1_2_10 is in error state because the deseq2 (1.2.10) dependency has this error :

* installing *source* package ‘XML’ ...
** package ‘XML’ successfully unpacked and MD5 sums checked
rm: cannot remove 'a.out.dSYM': Is a directory
ERROR: configuration failed for package ‘XML’
* removing ‘/home/galaxy/Formation/galaxy/dependencies/deseq2/1.2.10/iuc/package_deseq2_1_2_10/57b5f2997809/XML’

From what I see from the Internet, this error happens when the libxml2-dev is missing. I checked and I have it already on this Debian. Don't know if it really helps...

ADD REPLYlink modified 8 months ago • written 8 months ago by christophe.habib330

I've had ongoing headaches getting diffbind installed in Galaxy. If you can switch to using conda as a dependency resolver then that should make the installation easier.

ADD REPLYlink modified 8 months ago • written 8 months ago by Devon Ryan1.3k

This error is now solved thanks to this pull https://github.com/galaxyproject/tools-iuc/pull/892 . Now DiffBind has a beautiful green "Installed" status. But still it doesn't work and i still have the 2 errors mentioned before.

ADD REPLYlink modified 8 months ago • written 8 months ago by christophe.habib330
0
gravatar for j.m.fustin
8 months ago by
j.m.fustin10
j.m.fustin10 wrote:

Hi there,

It was quite a while ago but if I remember well I "resolved" the problem by simply accessing Galaxy and generating my files with a PC (English OS) instead of my MacBook, and it worked.

I hope it helps!

JM

ADD COMMENTlink written 8 months ago by j.m.fustin10

Hi Fustin,

Thank you for your answer.

Unfortunately I am using DiffBind with file generated within the Galaxy instance itself. I don't think I can apply your solution. I would rather have this tool to work properly.

I hope I will be able to find a solution.

ADD REPLYlink modified 8 months ago • written 8 months ago by christophe.habib330
0
gravatar for christophe.habib
8 months ago by
France
christophe.habib330 wrote:

I am trying to find to origin of this problem. I looked into uwsgi.log and in handler0.log or handler1.log. In this log I can see errors regarding unresolved dependencies for Rscript, DESeq2, and R. I already tried to uninstall everything related to DiffBind to install it again but it still doesn't work !

Here is what I see in the handler.log.

I can't find any further informations. The errors do not give many informations I think. Maybe I forgot a log file to check ?

Someone else meet this error ? My guess is that it is related to some system issue but I don't what to change to make things work.

ADD COMMENTlink written 8 months ago by christophe.habib330
0
gravatar for zachary.gaber
6 days ago by
United Kingdom
zachary.gaber60 wrote:

Has any progress been made on this issue? I tried to use diffBind today and had the exact same error message as reported here.

ADD COMMENTlink written 6 days ago by zachary.gaber60
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 75 users visited in the last hour