- galaxy-dev - lists.galaxyproject.org

Where is the xml file of "edit attribute"? and history name
by julie dubois 12 Nov '12

12 Nov '12

Hi, I've redirected all of my outputs in an other directory with my own names after each run of tools.. But when the users want to rename their datasets on the history, I would want that this change occurs on my output files already register in the computer. So where can I find the xml file of the function "edit attribute" and from this file, can I take the real name of my output file before the change? Another question is about the history name : because the users use several history, I need to know how call the variable of the history name ; for exemple if an output variable is $output, the history name of this variable is : $output.history_name ? or $output.history.name? or if this name is not accessible in this way, may be the history id ? In my first step when I rename an output, I take the $output.hid which indicate me the number of the dataset in an history. But I've not found the name of this history. Thanks Julie

1 0

How to get a compressed output in the history
by Philipe Moncuquet 11 Nov '12

11 Nov '12

Hi dev, I have written a perl script to perform some complex demultiplexing of reads. It works well on my computer but the integration to Galaxy is a bit tricky since it's the first time for me. At one point my perl script generates a bunch of fasta files, I want all this fasta files to be compressed and I use a tar command within the script to do so. I am not sure how to do the same within the Galaxy execution. I think it might work out with the "from_work_dir" attributes of the output tag but I don't know exactly where my script should write these fasta files and compress them so that the archive could be retrieve. I don't know either what to put in the "format" attributes since gz is not in the supported datatype or is it ? Should I create a new datatype ? The ending point of my tool would be to have the user be able to download one compressed file that would contains all the sorted fasta files. Cheers, Philippe

2 1

Uploading files onto local server
by John Clayton 09 Nov '12

09 Nov '12

Dear Sir or Madam, I am having difficult loading files onto my local galaxy instance. I have given myself admin status, but when I attempt to add files to a new library, I get the following message: Error attempting to display contents of library (New data library): (OperationalError) no such column: True u'SELECT dataset_permissions.id AS dataset_permissions_id, dataset_permissions.create_time AS dataset_permissions_create_time, dataset_permissions.update_time AS dataset_permissions_update_time, dataset_permissions.action AS dataset_permissions_action, dataset_permissions.dataset_id AS dataset_permissions_dataset_id, dataset_permissions.role_id AS dataset_permissions_role_id XnFROM dataset_permissions XnWHERE True AND dataset_permissions.action = ?' ['access']. I have tried toggling some things in the universe_wsgi.ini as well as changing the permissions associated with the library. But I still get this message. After looking around a bit on the help site, I cam across the following exchange on one of your boards: http://dev.list.galaxyproject.org/Error-while-uploading-data-from-local-dir… This is a bug that is fixed in revision ecd131b. Sorry for the inconvenience. -- jt However, I have updated to the most recent release using mercurial, so I don't know if this applies to me or not. How can I determine if I have revision ecd131b? I'm running Max Os 10.7 and using Safari to interface with galaxy. Actually, I was using Firefox at first, but it was hanging during file uploads and the problem disappeared when I switched to Safari. I guess this is a separate issue, but has anyone reported similar problems? -- John R. Clayton, PhD Université de Strasbourg Anopheles Group, CNRS UPR-9022 "Réponse Immunitaire et Développement chez les Insectes" Institut de Biologie Moléculaire et Cellulaire (IBMC) 15, rue René Descartes F-67084 CEDEX Strasbourg France Tel: +33 3.88.41.70.96 Fax: +33 3.88.60.69.22 Email: j.clayton(a)ibmc-cnrs.unistra.fr http://www-ibmc.u-strasbg.fr/ridi/index.php?cont_id=9&lang=en

2 1

Account had be deleted ?
by Diam Hsu 09 Nov '12

09 Nov '12

Dear Sir or Madam, I used to use Galaxy to analyze NGS data, but these two days I could't login my Galaxy account, it also told me that my account has been marked deleted, and asked me to contact the Galaxy administrator to restore my account. Could you help me restore my accounts because I have to use Galaxy to finish my thesis. Thank you very much for your help!!! Best regards, Dima Hsu Graduate student, Graduate Institute of Biomedical Informatics Taipei Medical University, Taipei, Taiwan Email:b1984073@gmail.com

2 1

which .loc file for "SAM to BAM"?
by Andreas Kuntzagk 09 Nov '12

09 Nov '12

Hi, I'm still setting up a local galaxy. Currently I'm testing the setup of NGS tools. If I try "SAM to BAM" for a BAM file that has "hg18" set as build I get a message that "Sequences are not currently available for the specified build." I guess that I have either to manipulate one of the .loc files (but which?) or have to download additional data from rsync server. (I already have the tool-data/shared/hg18 completely) regards, Andreas Btw. Do you have any plans to ease the pain on adding additional builds? Something simpler than having to add one line for each build*tool combo? These lines seem very redundant to me. -- Andreas Kuntzagk SystemAdministrator Berlin Institute for Medical Systems Biology at the Max-Delbrueck-Center for Molecular Medicine Robert-Roessle-Str. 10, 13125 Berlin, Germany http://www.mdc-berlin.de/en/bimsb/BIMSB_groups/Dieterich

3 10

GATK fails to create Java Virtual Machine
by Iry Witham 09 Nov '12

09 Nov '12

Hi Folks, I received an email from one of my end users that she was having an issue with the "Realigner Target Creator" under NGS: GATK Tools. Here's the error: An error occurred running this job: "Unable to access jar file /hpcdata/galaxy-setup/galaxy-dist/tool-data/shared/jars/gatk/GenomeAnalysisTK.jar." That was easy to resolve since the jar file is located at either /hpcdata/galaxy-setup/galaxy-dist/tool-data/shared/jars/GATK/1.4.2/ or /hpcdata/galaxy-setup/galaxy-dist/tool-data/shared/jars/GATK/1.6.13/. I modified the xml to point to the /1.4.2 directory and now I get "Could not create the Java virtual machine." I have looked at the wrapper for java settings and restarted the Galaxy server, but am at a loss. Do you have any ideas? Thanks, Iry Witham Scientific Applications Administrator Scientific Computing Group Computational Sciences Dept. The Jackson Laboratory 600 Main Street Bar Harbor, ME 04609 Phone: 207-288-6744 email: iry.witham(a)jax.org The information in this email, including attachments, may be confidential and is intended solely for the addressee(s). If you believe you received this email by mistake, please notify the sender by return email as soon as possible.

1 0

makeblastdb wrapper
by Espin, Almudena 09 Nov '12

09 Nov '12

Hi guys, I am writing a wrapper for makeblastdb. The problem is that Galaxy doesn’t support the blastdb data format, so this has to be added if a makeblastdb wrapper is to work properly. I have problems modifying the registry.py, Could anyone specify which changes have to be done? Thank you

2 3

How to get a compressed output in the history for my own tool
by Philipe Moncuquet 09 Nov '12

09 Nov '12

Hi dev, I have written a perl script to perform some complex demultiplexing of reads. It works well on my computer but the integration to Galaxy is a bit tricky since it's the first time for me. At one point my perl script generates a bunch of fasta files, I want all this fasta files to be compressed and I use a tar command within the script to do so. I am not sure how to do the same within the Galaxy execution. I think it might work out with the "from_work_dir" attributes of the output tag but I don't know exactly where my script should write these fasta files and compress them so that the archive could be retrieve. I don't know either what to put in the "format" attributes since gz is not in the supported datatype or is it ? Should I create a new datatype ? The ending point of my tool would be to have the user be able to download one compressed file that would contains all the sorted fasta files. Cheers, Philippe

1 0

Testing framework and repeat, boolean and conditional parameters
by Simon Gladman 09 Nov '12

09 Nov '12

Hi all, Have looked through the list archives regarding this without much luck. I have a repeat parameter which contains a boolean based conditional. I wrote a simple test case for it but I think the test keeps failing at the boolean parameter.. input section looks like: <repeat name="inputs" title="Input read libraries"> <conditional name="is_separate"> <param name="separate" type="boolean" checked="false" truevalue="-separate" falsevalue="" label="Are the reads paired and in two different files?" /> <when value=""> <param name="forwards_file" type="data" format="fasta,fastq,raw,sam,bam" label="Read dataset"/> </when> <when value="-separate"> <param name="forwards_file" type="data" format="fasta,fastq,raw,sam,bam" label="Read dataset for direction 1"/> <param name="reverse_file" type="data" format="fasta,fastq,raw,sam,bam" label="Read dataset for direction 2"/> </when> </conditional> ....... </repeat> test section looks like: <tests> <test> <param name="start_kmer" value="21"/> <param name="end_kmer" value="29"/> <param name="inputs.is_separate.separate" value="false"/> <param name="inputs.is_separate.forwards_file" value="vopt_test_reads.dat" ftype="fasta"/> <output name="contigs" file="vopt_test_contigs.dat"/> </test> </tests> The output from run_functional tests is: ====================================================================== ERROR: test_tool_000000 (functional.test_toolbox.TestForTool_velvetoptimiser_vlsci) Velvet Optimiser vlsci ( velvetoptimiser_vlsci ) > Test-1 ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/simon/Galaxy/galaxy-dist/test/functional/test_toolbox.py", line 171, in test_tool self.do_it( td, shed_tool_id=shed_tool_id ) File "/home/simon/Galaxy/galaxy-dist/test/functional/test_toolbox.py", line 76, in do_it page_inputs = self.__expand_grouping(testdef.tool.inputs_by_page[0], all_inputs) File "/home/simon/Galaxy/galaxy-dist/test/functional/test_toolbox.py", line 131, in __expand_grouping expanded_inputs.update( self.__expand_grouping( { new_prefix : r_value }, declared_inputs, prefix = new_prefix ) ) File "/home/simon/Galaxy/galaxy-dist/test/functional/test_toolbox.py", line 118, in __expand_grouping if declared_inputs[ value.test_param.name ] == case.value: KeyError: 'separate' -------------------- >> begin captured stdout << --------------------- kmer_step not declared in tool test, will not change default value. filetype not declared in tool test, will not change default value. --------------------- >> end captured stdout << ---------------------- -------------------- >> begin captured logging << -------------------- galaxy.web.framework: DEBUG: Error: this request returned None from get_history(): http://localhost:8326/ galaxy.web.framework: DEBUG: Error: this request returned None from get_history(): http://localhost:8326/ galaxy.web.framework: DEBUG: Error: this request returned None from get_history(): http://localhost:8326/user/logout galaxy.web.framework: DEBUG: Error: this request returned None from get_history(): http://localhost:8326/ --------------------- >> end captured logging << --------------------- ---------------------------------------------------------------------- Ran 1 test in 0.675s FAILED (errors=1) My question: Can the testing framework handle the nesting of the boolean, conditional and repeat parameters or have I made a mistake in the test xml? Cheers, Simon Gladman VBC/VLSCI Monash University Australia.

1 0

Velvet not working
by Philipe Moncuquet 09 Nov '12

09 Nov '12

Hi developers, We have our own instance running and under continuoius development. I installed velvetg and velveth from the toolshed (revision 0:4afe13ac23b6<http://toolshed.g2.bx.psu.edu/repository/view_changelog?id=9dca5b8f68c509af>) and manually installed binaries for velvet_1.2.08. When I try to test it under the galaxy web interface, the job does not failed but all files are empty except the first one that contains the following error message : [0.000000] Unknown option: -ins_length3 I guess this might come from the version of the binaries and the version the xml wrapper was written for are different. Does anyone knows which binaries version I should use ? How is it that the job does not appears as failed ? Any help will greatly appreciated. Regards, Philippe

3 3