Question: Galaxy fatal error running Cutadapt
0
gravatar for gkuffel22
7 months ago by
gkuffel22170
United States
gkuffel22170 wrote:

Hi,

I manage our private instance of Galaxy on Jetstream. I am having all kinds of issues installing new tools. I have made sure to enable Conda but I haven't been able to successfully install many tools. One example would be with CutAdapt. It looked like it installed just fine but when I run the tool I get this error:

ERROR: Running in parallel is currently not supported for the given combination of command-line parameters.
These options are not supported: --info-file, --rest-file, --wildcard-file, --untrimmed-output, --untrimmed-paired-output, --too-short-output, --too-short-paired-output, --too-long-output, --too-long-paired-output, --format, --colorspace

Can anyone help so that I can figure out how to actually manage my Galaxy instance? Thanks in advance.

ADD COMMENTlink modified 7 months ago by mblue9140 • written 7 months ago by gkuffel22170
2
gravatar for mblue9
7 months ago by
mblue9140
mblue9140 wrote:

Hello,

Sorry to hear you're having problems with cutadapt. I was the one who updated the Galaxy wrapper recently and we have it installed in 17.09, I've only used it a little bit so far but it seems to be working fine. But note that there are currently some limitations with cutadapt's parallel/multi-core e.g. not all the outputs can be produced http://cutadapt.readthedocs.io/en/stable/guide.html#multi-core-support

Btw my strategy for troubleshooting tool issues like this is to first doublecheck that the test data runs ok.

For example:

Take one of the test data files from the tool's test-data folder, as they're small so will be quick to run, like this one: https://github.com/galaxyproject/tools-iuc/blob/master/tools/cutadapt/test-data/bwa-mem-fastq1.fq.gz

Then run it through the tool using the same parameters as in the test e.g. see here (also pasted below): https://github.com/galaxyproject/tools-iuc/blob/master/tools/cutadapt/cutadapt.xml#L331-L338

<!-- Ensure single end fastq.gz works --> 
<test>
    <param name="type" value="single" />
    <param name="input_1" ftype="fastq.gz" value="bwa-mem-fastq1.fq.gz" />
    <param name="adapter_source_list" value="user"/>
    <param name="adapter" value="AGATCGGAAGAGC"/>
    <output name="out1" decompress="True" file="cutadapt_out1.fq.gz"/> 
</test>

Hope that helps!

P.S. Thanks Jen for passing on the notification of this in Gitter.

ADD COMMENTlink written 7 months ago by mblue9140
0
gravatar for Jennifer Hillman Jackson
7 months ago by
United States
Jennifer Hillman Jackson25k wrote:

Hello,

Note: This is a continuation of the troubleshooting in this post: https://biostar.usegalaxy.org/p/27205

I've asked the tool developers for help directly at Gitter: https://gitter.im/galaxy-iuc/iuc?at=5acbb20e92f5d62057ae0d60

They may answer back here or you can continue the troubleshooting at Gitter.

For reference, Galaxy Administrator help can be found in these places:

Thanks! Jen, Galaxy team

ADD COMMENTlink modified 7 months ago • written 7 months ago by Jennifer Hillman Jackson25k

Hi Jen,

Thanks for the input. Actually I have two separate instances of Galaxy, one on Jetstream and one running on our servers through a virtual machine. I can't control the version of the instance on Jetstream but we have the newest version running on our servers. Point is, I have different problems with each instance. I cannot install any tools on either instance, the dependencies are not being handled correctly eventhough I have tried everything.

If I share my password info with you would you be willing to take a look at the .yml file and how I have set up the path to Conda? I am probably missing something that is obvious. Thanks so much.

ADD REPLYlink written 7 months ago by gkuffel22170

Please share your configuration with the tool author and developers to troubleshoot. Do this through Gitter. You don't need to share your password, just state that you need configuration help and share your file (redact anything that you do not want to be public). Share file contents with a link: either put your file someplace public and share the URL or paste the contents into an online utility like Gist https://gist.github.com/ (the second is often better, especially if only certain lines need to be reviewed).

When issues like this come up there is generally some back and forth to resolve the problem, so a direct chat is quicker with fewer time delays in the conversation.

ADD REPLYlink written 7 months ago by Jennifer Hillman Jackson25k
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: 173 users visited in the last hour