Oncotator integration with galaxy

hcountouhcountou FranceMember

Hello,

i'm working on integrating tools to a galaxy plateform for cancer research, based on whole sequencing data.

Thus a question rise "suddenly". People here are used to perform task with a R script on the distant oncotator site. But this process should be internalized for obvious internal secure reasons.

So my question is : should i integrate all scripts from source (from the oncotator package i mean) in Galaxy or GATK, provided as a galaxy wrapper, has functions that are exactly the same as those included in this Oncotator package (i don't think so, but...) for annotation.

Has anyone tried that before ? Sorry if i ask in the wrong place.

Thanks in advance for your help.

Best Answer

Answers

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    @hcountou I'm sorry but I don't understand what you are asking. Can you please clarify?

  • hcountouhcountou FranceMember

    Hello,

    sorry for not being clear enough.

    People here are using the oncotator annotation through a R script. This is not compatible anymore with our security policy.

    Now i have to integrate this process into our Galaxy plateform anyway.

    So it seems i have 3 choices
    1 - leave it has it is and just build a wrapper for this R script in Galaxy
    2 - install all oncotator scripts that you provide including the db files (download links on the oncotator install web page) and build wrappers again for each function
    3 - Use GATK instead of oncotator for annotation, but then i have to be sure it'll be the exact same result.

    Can you please give me an advice on what should be the best way to achieve that oncotator integration into galaxy.

    Hope my explanations are a little bit better.
    Thanks in advance for your help.

  • hcountouhcountou FranceMember

    Ok. Thanks for your help. I confirm our goal is a non-profit one. Our work is to improve our hospital translational platform and so our data analysis abilities. If we success to adapt oncotator for galaxy, i'll let you know and we'll share with the community of course.

    Best regards and thank you again.

  • hcountouhcountou FranceMember

    Hello,

    it seems i'm on the right way. Oncotator and db files have been installed and are working. I can call and produce annotation files from either tabular file or VCF file. But...

    I have a problem because the script is producing a lot of "INFO" message and that is understood by the Galaxy server as errors. This actually raise an error status despite the fact that files are correctly produced (i add a supplementary log file as output product containing those "INFO"/"WARNING" messages).

    I tried but it seems i can't turn those message off by using the "-v" option set to "0" or "ignore" status. I don't want to edit the script if possible. Do you have any clue ?

    Thanks in advance for your help.

    Best regards.

  • hcountouhcountou FranceMember

    Sorry,

    i don't really find a proper solution and this is one of my last problem to solve. I need to know how to shut down le "verbose mode on" that is launched automatically when i execute a local oncotator annotation. Do you have any clue ?

    Verbose mode on
    Path:
    ['/usr/local/python2.7/lib/python2.7/site-packages/Oncotator-v1.5.1.0-py2.7.egg',..., '/home/galaxy/.local/lib/python2.7/site-packages']

    Best regards ?

  • hcountouhcountou FranceMember

    Ok it seems that what i need is to remove the console logger that you add on purpose because Galaxy treats all message as errors on the running script. Any idea how to do so ? The argument "-v 0" is not working properly.

  • hcountouhcountou FranceMember

    I've found a solution (i mean a suitable one for our Galaxy operational needs) :

    1 - Force verbose mode to 0
    parser.add_argument("-v", "--verbose", dest="verbose", action="count", help="set verbosity level [default: 5]", default=0)

    I don't understand why the default value is 5 by the way ? But ok let's change it.

    2 - Comment all the following coding block in order to remove all console messages (WARN are still active, but is that all ?) :

       # Add a console logger to the root logger, which means that all loggers generated will have the console dump. 
        #    Output on the console will be the same as what is in the log file.
        # ch = logging.StreamHandler()
        # ch.setLevel(logging.WARN)
        # formatter = logging.Formatter(loggingFormat)
        # ch.setFormatter(formatter)
    
        # if verbose:
        #    ch.setLevel(logging.INFO)
        #    print("Path:")
        #    print(sys.path)
        #    print(" ")
    
        # logging.getLogger('').addHandler(ch)
    
        # logger = logging.getLogger(__name__)
        # logger.info("Oncotator " + program_version)
        # logger.info("Args: " + str(args))
        # logger.info('Log file: ' + os.path.abspath(logFilename))
        #
        # if DEBUG:
        #     logger.setLevel(logging.DEBUG)
    
        # if not NGSLIB_INSTALLED:
        #     logger.warn("ngslib module not installed.  Will be unable to annotate with BigWig datasources.")
    

    Note : watch out wich version or egg you're modifying (2.6 or 2.7). We run Oncotator script with the recommended 2.7 version of python.

  • hcountouhcountou FranceMember

    One more thing. Commenting the block for the "console logger" implementation doesn't mean you don't get the log infos, but you only have them in a separated file and not in your prompt.

Sign In or Register to comment.