Cmake error while bcbio-nextgen installation about bcbio-nextgen HOT 15 CLOSED

Shalabh;
Thanks for the problem report. It looks from the error log that there is something problematic about the cmake install on your system, which is causing the bamtools build to fail. How have you installed cmake? What system are you working on? Would it be possible to do a reinstall of cmake and re-run?

from bcbio-nextgen.

It looks like a missing C/C++ compiler to me. If you are using Ubuntu/Debian install build-essential, if you are using openSUSE install "gcc" and "gcc-c++", if you are using Fedora / RHEL install "gcc" and "gcc-c++".

from bcbio-nextgen.

Thank you for your response. I was able to resolve the issue by reinstalling cmake version 2.8 and exporting correct paths for g++ and gcc

But after rectifying the above issue and going further down with the installation, I am having trouble during the Import leiningen step. The script seems to get struck (...without any activity for hours...) during this step without any error message. Here is a screenshot:

Please suggest accordingly.

PS - I am working on a CentOS 6.4 and running the following command inside Cloudbiolinux directory:
fab -f ./fabfile.py -c ./contrib/flavor/ngs_pipeline/fabricrc_custom_071713.txt -H localhost install_biolinux:flavor=ngs_pipeline

Thank you

Shalabh Suman

from bcbio-nextgen.

Shalabh;
Glad you're getting further along and thanks for the feedback about the problem. I pushed fixes to the CloudBioLinux install to avoid some of the problematic packages, including leiningen. If you update your cloudbiolinux from GitHub, it will skip this step and hopefully work cleanly for you. Please let us know if you run into other issues.

from bcbio-nextgen.

Thank you for the response.
Is there a command to efficiently update existing cloudbiolinux install? Or should I remove and reinstall the complete cloudbiolinux package?

Thanks

Shalabh Suman

from bcbio-nextgen.

Shalabh;
You can do git pull origin master from inside the cloudbiolinux directory if you originally got it by cloning from GitHub. If not, then remove and redownload is the best path.

Also, the latest version has a more minimal flavor you can use by substituting ngs_pipeline_minimal for ngs_pipeline in the command line. That will have fewer packages but enough to get started with bcbio-nextgen.

Hope this helps.

from bcbio-nextgen.

Thank you very much. I have another question:

If I would like to use PYTHON 2.7 instead of default python 2.6 for installing packages inside cloudbiolinux directory, what would be the recommended way to accomplish this? Should I be make changes inside any of the main.yaml files for the same?

Thanks

-Shalabh Suman

from bcbio-nextgen.

Shalabh;
If you add:

python_version_ext = 2.7

to confg/fabricrc.txt this will use 2.7 instead of the 2.6 addition. You'll need to update to the latest version I just checked in to avoid code that overwrites this. What it's trying to do on CentOS is avoid using the old 2.4 default version. Hope this works for you.

from bcbio-nextgen.

Thank you for all your help. It seems finally I have been able to accomplish the installation with no more error messages.

However I was trying to run one of your example pipelines called Exome with validation against reference materials

• using this command:
  bcbio_nextgen.py /usr/local/variant_caller_ensemble/ensemble_test_071513/bcbio-nextgen/config/bcbio_system.yaml ../input/ ../config/NA12878-exome-methodcmp.yaml -n 8
• and I am getting the following error:
  

Do you have any suggestion?

Thanks

Shalabh Suman

from bcbio-nextgen.

Shalabh;
The pipeline uses the location of bcbio_system.yaml to find the galaxy directory with the *.loc files. It looks like you copied it over into the configuration directory, so it doesn't know where to find those files. Your best bet is to leave bcbio_system.yaml in the original location from the installer. You can symlink it in if that is easier to remember.

Sorry for the confusion. This is tricky but I've not yet come up with a better solution that doesn't require even more configuration.

from bcbio-nextgen.

Thank you for your response.

However, I am a bit confused about the ORIGINAL location of bcbio_system.yaml?? The reason I am asking this is because, after I manually installed bcbio-nextgen using following commands:

git clone https://github.com/chapmanb/bcbio-nextgen.git
cd bcbio-nextgen && python setup.py build && sudo python setup.py install

I see bcbio_nextgen.py inside /usr/local/bin and /bcbio-nextgen/scripts directory, but bcbio_system.yaml only inside /bcbio-nextgen/config/ directory.

I guess I am doing it correctly??

Thanks

Shalabh Suman

from bcbio-nextgen.

Shalabh;
Have you installed any of the associated data files? That process sets up a galaxy directory with correct *.loc files pointing to installed genome builds.

The automated installer takes care of this:

https://bcbio-nextgen.readthedocs.org/en/latest/contents/installation.html#automated

or you can manually run it:

https://bcbio-nextgen.readthedocs.org/en/latest/contents/installation.html#data-requirements

I'll have a new version of the pipeline and installer out next week that should make the automated installation and data updates easier.

Once you have the data directory, the easiest way to set it up is to place bcbio_system.yaml into the galaxy directory and it'll then be able to find all of the associated *.loc files.

from bcbio-nextgen.

Thank you for your response. I did already install the associated data files, but somehow the galaxy directory was missing. So, after reinstalling it and placing bcbio_system.yaml into it, I ran the following command to test the same example pipeline called Exome with validation against reference materials :

bcbio_nextgen.py /usr/local/variant_caller_ensemble/biodata/galaxyTools/galaxy-central/bcbio_system.yaml /usr/local/variant_caller_ensemble/ensemble_072313/input/ /usr/local/variant_caller_ensemble/ensemble_072313/config/NA12878-exome-methodcmp.yaml -n 8

and I get the following error message (total error in two screen shots):

I really appreciate your continual help on this.

Shalabh Suman

from bcbio-nextgen.

Shalabh;
Great work getting this all set up. It looks like you've got it generally running smoothly.

To run this specific example you'd need to have a licensed version of novoalign, since only licensed versions handle gzipped input files and parallelization. You can edit the NA12878-exome-methodcmp.yaml to remove the blocks that use novoalign and this will run only the bwa sections and is still a good exercise of the system. I'll add this information to the documentation and will work on making a simpler first example to use that only uses bwa. Thanks much.

from bcbio-nextgen.

Shalabh;
Closing this one now to clean up after the new release. If you run into any other issues please let me know. Thanks for all this feedback.

from bcbio-nextgen.