April 2011 - galaxy-dev - lists.galaxyproject.org

PostgreSQL issue: reserved keyword used as identifier
by Louise-Amélie Schmitt 09 Apr '11

09 Apr '11

Hello everyone I just met a huge problem concerning the database. I'm currently trying to transfer my data from MySQL to PostgreSQL by writing a Perl script that would do the job. Here is the issue: In the "form_definition" table, one of the field identifiers is "desc", which is a reserved SQL keyword used for ordering values. Therefore, There's currently no way of making any query of the type "INSERT INTO table_name (<identifiers list>) VALUES (<values list>);" which is a big handicap in this context, since the order of the identifiers list we dynamically retrieve is not necessarily (and seldom) the right order. Is there a way to fix this issue without blowing everything up? Cheers, L-A

4 6

April 8, 2011 Galaxy Development News Brief
by Jennifer Jackson 08 Apr '11

08 Apr '11

April 8, 2011 Galaxy Development News Brief http://bitbucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011_… Prior releases: https://bitbucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBriefs ------------ 2011_04_08 Highlights Workflow API Move Data Library Items 10 New Genomes & 29 New LiftOver Import Workflows & Histories ClustalW Multiple-Seq Alignments & Sequence Logos Major Trackster Visualization Improvements ------------ How to get this distribution https://bitbucket.org/galaxy/galaxy-dist/ new: % hg clone http://www.bx.psu.edu/hg/galaxy galaxy-dist upgrade: % hg pull -u -r 50e249442c5a ------------ Key Upcoming Galaxy Events May 24-26, 2011 Galaxy Community Conference, Lunteren, The Netherlands. http:galaxy.psu.edu/gcc2011/ April 24 is the Early registration deadline (save 20%!) http://galaxy.psu.edu/gcc2011/Register.html April 13-14, NBIC Galaxy Hackathon, Belgium. Please submit your suggestions! http://wiki.nbic.nl/index.php/NBIC_Galaxy_Hackathon_project ------------ What's New Workflow API An API for executing workflows has been added. Usage examples /scripts/api/workflow_execute.py /scripts/api/example_watch_folder.py. Move Data Library Items We have introduced a feature that enables you to move library datasets or folders (including all folder contents) to other locations within the same data library or to a different data library altogether. Galaxy administrators can perform this feature on any data library item and can move items to any data library. Users that are not Galaxy administrators must be given the modify library item permission on an item in order to move it, and they'll need the add library item permission on the desired target data library folder in order for it to be displayed in the select list of targets. Moving a single dataset To move a single dataset, select the Move this dataset option from the dataset's pop-up menu. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… The default behavior is to move items to other locations within the same data library, so you'll initially be presented with a list of valid folders from which to choose for the new target location. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… After selecting the desired folder, clicking the Move button will move the dataset to the selected folder. Moving a Folder Moving a folder is very similar to moving a single dataset - just select the Move this folder option from the folder's pop-up menu. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… To move the folder to a different data library, after selecting the Move this folder option, click the Choose another data library link on the Move data library items page. The folders select list will change to a select list of data libraries to which you are authorized to move the item. When you select a target data library, the select list will change again to display the list of folders in the selected data library to which you are authorized to move the item. Clicking the Move button will move the folder and all of it's contents to the selected target folder within the selected target data library. The target folders select list is filtered to included only valid folders to which you can move the item. For example, you cannot move a folder to one of it's own sub-folders in one step. To do this, the sub-folder must be moved outside of it's parent, and then the parent can be moved to the folder that it previously contained. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… ------------ Updated & Improved Data Content * Genomes o New & included in NGS Tools + Saccharomcyes cerevisiae: Saccharomcyes_cerevisiae_S288C_SGD2010 + Arabidopsis lyrata: Araly1 + Purple Sea Urchin: strPur3 and Spur_v2.6 + Hydra: Hydra_JCVI + Zebrafish: danRer7 + Poplar: Ptrichocarpa_156 + Chimpanzee: panTro3 + Northern White-Cheeked Gibbon: nomLeu1 + Korean Man AK1: Homo_sapiens_AK1 o New & not included in NGS Tools + Caenorhabditis remanei: caeRem2 o Existing genomes added to NGS Tools + hg19 Canonical female (no Y chromosome) + Streptococcus pneumoniae R6: 278 + Drosophila virilis: droVir3 and droVir2 * New LiftOver Files o caeRem2 --> caePb1, caeRem2 --> caeRem3, caeRem2 --> cb3, caeRem2 --> ce4, caeRem2 --> priPac1, calJac3 --> hg18, canFam2 --> monDom5, danRer6 --> danRer7, danRer7 --> fr2, danRer7 --> gasAcu1, danRer7 --> hg19, danRer7 --> mm9, danRer7 --> oryLat2, danRer7 --> panTro3, danRer7 --> tetNig2, danRer7 --> xenTro2, droVir3 --> droVir2, fr2 --> danRer7, gasAcu1 --> danRer7, hg18 --> calJac3, hg19 --> danRer7, mm9 --> danRer7, panTro3 --> danRer7, panTro3 --> hg19, ponAbe2 --> calJac3, ponAbe2 --> monDom5, strPur2 --> ci2, tetNig2 --> danRer7, xenTro2 --> danRer7 * Add Genomes to Your Instance http://bitbucket.org/galaxy/galaxy-central/wiki/NGSLocalSetup * Current Galaxy Main Genomes http://bitbucket.org/galaxy/galaxy-central/wiki/GenomeData Current Tools * Add more verbose error reporting to FASTQ Groomer tool. Provides more information to allow users to determine what is wrong with FASTQ files with invalid format. * Enhance Bowtie wrapper to accept non-Sanger variant FASTQ files. * Allow Upload Tool to function on https URLs. * Add count GFF features tool and tests: o Filter and Sort --> GFF --> Filter GFF file by feature count using simple expressions. o Tool counts the number of features in a GFF file. Note: this is different than the number of lines because a single GFF feature can often span multiple lines. * Tophat v1.2.0 support: o (a) allow indel search. o (b) max insertion and max deletion lengths. o (c) library type. * Updated gff_filter_by_feature_count tool now accepts and correctly handles all GTF, GFF, and GFF3 files. * Changes for detecting and loading BAM data type Samtools version 0.1.13 or newer produces an error condition when attempting to index an unsorted BAM file. To determine if a BAM file is sorted, we first use Samtools to check the headers. If this does not provide a definitive answer and Samtools version 0.1.13 or newer is being used, we index the BAM file to see if it produces the error. This process provides a more robust approach to determining if the BAM file is sorted. New Tools * Multiple Alignments: ClustalW multiple sequence alignment program for DNA or proteins. * Motif Tools: Sequence Logo generator for FASTA data (example: ClustalW alignment). o Both tools originated from the Community Tool Shed (see below). o The Sequence Logo tool uses Weblogo3 wrapped into Galaxy to generate a sequence logo. The input file must be a FASTA file in your current history. It is recommended for viewing multiple-sequence alignment output from the ClustalW tool. Set the ClustalW output to FASTA to create the input for this tool. A typical output looks like this: http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… Community Tools (Tool Shed) * Tuning: Clarified what is being searched in the Tool Shed. * New ClustalW wrapper (see http://www.clustal.org) for protein/dna multiple alignments based on the Galaxy ClustalW wrapper posted by Hans-Rudolf Hotz in an email on the developer list. http://lists.bx.psu.edu/pipermail/galaxy-dev/2010-November/003732.html * New Weblogo3 wrapper (http://weblogo.berkeley.edu) that creates sequence logos from FASTA data such as the output from a ClustalW alignment. Data Libraries * Disabled problematic eager loading on data libraries. Very large data libraries will load two to three times quicker. * Upload Improvement with example walk-through: Uploading data library datasets 1. Apply a corrected version of the patch from Peter Cock that flips the "Preserve directory structure?" setting when uploading library datasets from filesystem paths. The checkbox is now "yes" instead of the original "no", and is checked by default. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… 2. Flip the behavior of the "Copy data into Galaxy" feature when uploading library datasets (similar to 2 above) to be more clear and logical. Instead of a single checkbox, this is now a select list with clearly defined options. The default setting is to copy the files into the configured Galaxy file store. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… 3. Allow importing items from a history to replace a library dataset with a new version. Previously, you could only replace a library dataset with a new version by uploading a single file. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… Workflows * Users can now import copies of their own Workflows and Histories. Trackster * Enhancements: o Implemented a data manager for Trackster and drawing is now completely tracked and controlled. o Put show_insertions and show_differences in read track config. o Filtering now supports: + (a) score columns in BED and GFF. + (b) GTF attributes. o Added BED and GFF support to visual analytics framework, and enable GOPS intersect and subtract tools to work with visual analytics. * Improvements: o Insertions and deletions now shown in reads. o Save and restore mode for feature and read tracks. o Bug fixes for drawing feature tracks in Squish and Dense mode. o Menus and menu items now work correctly. o Remove form from navigation controls so that enter key works properly and sets the chrom/low/high. o Enable full keyboard navigation via arrow keys. * Bug fixes: o Fix issues with navigation input: + (a) arrow keys no longer perform navigation + (b) invalid chromosome names are handled well. o Fix feature track bugs so that intervals are correctly drawn as half-open. o Use gap to correctly position read connector in Pack mode. o Fix read id bug in Trackster BAM data provider. Example of Trackster visualization: Track of reads mapped using Tophat (spliced) including the transcript assembled from those reads. http://bytebucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011… User Interface (UI) * SGD Yeast genome added as a Gbrowse display site. * IGV added as an external display application. This is not enabled on Galaxy main, but is available for local instances. CloudMan * FTP data upload enabled. http://bitbucket.org/galaxy/galaxy-central/wiki/cloud Source * Improvements in reliability and speed regarding manipulation of instance data volumes. * The value set in 'new_file_path' in universe_wsgi.ini will now be used for creation of all temporary files. This obviates the need for setting $TEMP when running data source tools on the cluster. * Datatypes: o Add a 'subclass' flag to datatype definitions in datatypes_conf.xml that allows dynamic creation of a subclass. o Allow composite datatype datasets to be populated in the Upload tool from files that were uploaded to the Galaxy FTP server. * Tool configuration enhancements: o Allow DataToolParameter to be filtered on attributes other than dbkey (dynamic options). o Allow FromParamToolOutputActionOption and ParamValueToolOutputActionOptionFilter to access attributes of parameter values. Tool Test Framework * Specifying 'ftype="bam"' as a parameter on a test's <output> tag will now cause the test framework to use 'samtools view' to convert the file to SAM. This should make it easier to debug why the test tool output differs from your baseline test output. Test framework enhancements: * Allow toolbox tests to upload a file found located in subdirectories. * Fix a bug occurring on the determination of uploaded dataset name during the handling of the removal of .gzip or .zip extension from the uploaded filenames. * When using re_match comparison method in functional tests and line counts do not match, print out first 40 lines of the history file. Admin Menu * Add a checkbox to the Create Group page that if checked will create a new Role with the same name. This provides a similar feature the the existing checkbox on the Create Role page. * However, the behavior is now changed such that new associations are created when the checkbox is checked, whereas before, only the Group or Role objects with the same name were created, but not associated with anything. Bug Fixes * Following the login link from the Logout page no longer redirects you to the Logout page once you have logged in. * The peek setting method, set_peek(), should now function consistently across datatypes descending from Text, particularly with respect to line count estimation. * Apply patch from Ry4an Brase to correct "NoneType preference on the jobs view" issue. * Don't alter the contents of a file while uploading to a data library if using one of the Upload files from file system paths or Upload a directory of files options in conjunction with the Link to files without copying into Galaxy option. This partially resolves the issue where a supposedly sorted BAM file was being resorted upon upload to a data library when using this option. A better implementation of determining whether a BAM file has been sorted (so that it does not get resorted) remains to be done. * Extract Genomic DNA tool: o Do not fix strand for GFF input. o Handle non-GFF files when interpret features is true. * Set default value for Cuffdiff's minimum alignment count parameter to reflect v0.9.3. * Enhance GFF reader to handle headers and comments and fix bug so that new feature starts when transcript_id or gene_id changes. * Suppress most logging for Cufflinks and Cuffdiff as the logging can generate output that is incompatible with UTF-8 format used by database. * Set message type to error when history deletion fails due to sharing. * Show masthead links to Published Visualizations and Published Pages only if Tracks or Pages are enabled. * Shared visualizations work again. * Fix bug in recently used tool submenu. * Change default value for Tophat coverage search to true. ------------ About Galaxy The Galaxy team is a part of BX at Penn State, and the Biology and Mathematics and Computer Science departments at Emory University. http://www.bx.psu.edu/|BX/ http://www.psu.edu/ http://www.emory.edu/ http://www.biology.emory.edu/ http://www.mathcs.emory.edu/ Galaxy is supported in part by NSF, NHGRI, the Huck Institutes of the Life Sciences, and The Institute for CyberScience at Penn State, and Emory University. http://www.nsf.gov/ http://www.genome.gov/ http://www.huck.psu.edu/ http://www.ics.psu.edu/ http://www.emory.edu/ Source http://bitbucket.org/galaxy/galaxy-central Use Galaxy! http://usegalaxy.org GalaxyProject http://galaxyproject.org Join us at Twitter #usegalaxy http://twitter.com/#!/search/galaxyproject ------------

1 0

reason/advantages for making Galaxy proxy using mod_rewrite instead of mod_proxy
by Leandro Hermida 08 Apr '11

08 Apr '11

Hi all, Sorry if this question has been asked before searched and couldn't find anyone specifically talking about this. In the Galaxy wiki under https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ApacheProxy there are instructions for setting up Galaxy behind a web server proxy. I wondered what are the reasons/advantages of doing a reverse proxy using all mod_rewrite directives (as shown in the instructions) vs doing a standard Apache front-end reverse proxy: ProxyRequests Off ProxyPass / http://galaxy.host.domain:8080/ ProxyPassReverse / http://galaxy.host.domain:8080/ The instructions using mod_rewrite are making a reverse proxy too since the RewriteRule directive has the [P] flag. regards, Leandro

2 3

Trouble viewing interval files in UCSC with Apache+XSendFile
by Assaf Gordon 07 Apr '11

07 Apr '11

Hello, I've encountered a strange combination of factors that results in file access problems. Perhaps I'm doing something wrong - any advice will be appreciated. I'm using Apache + XSendFile. when viewing a strict BED file, everything works, because the actual dataset filename is passed on to Apache/XSendFile. When viewing any other kind of interval file, Galaxy first creates a temporary file in strict BED format ( in "Interval::as_ucsc_display_file()" ). This method uses "tempfile.mkstemp()", which is documented to create a temporary file with user-only read/write access (no group or world access). So when the file name is passed on to apache/XsendFile - apache can't read the file and returns 404. I'm wondering if anyone else encountered such problem, or is my configuration is somehow wrong. Thanks, -gordon

3 4

samtools sam-to-bam problem
by Ryan Golhar 07 Apr '11

07 Apr '11

I've performed an alignment using BWA on a file of paired-end illumina reads. The SAM file looks fine, and contains header information. I'm converting it to BAM using the sam to bam converter, however it consistently errors out after running for a while. The error is: "Error extracting alignments from (/home/galaxy/galaxy-dist/database/files/000/dataset_785.dat), " but no error is provided. Looking at the sam_to_bam.py on line 156 is where the error is thrown. Nothing is in e (I think). BTW - If I run the samtools command from the shell by hand, the BAM file is created properly. I do see information on stderr: $ samtools view -bt /data/genomes/H_sapiens/hg19/hg19.fa.fai -o /tmp/killme.bam /home/galaxy/galaxy-dist/database/files/000/dataset_785.dat [samopen] SAM header is present: 25 sequences. I'm using samtools version 0.1.14 (r933:170) on Linux, 64-bit. What do I do? Ryan

4 14

Enacting Taverna2 workflows through Galaxy
by Kostas Karasavvas 07 Apr '11

07 Apr '11

Hi all! I would like to bring to your attention a project that aims to make Taverna and Galaxy work better together. This new application is a first step towards making Galaxy and Taverna2 (http://www.taverna.org.uk) more interoperable. It allows for the automatic generation of a Galaxy tool from a Taverna2 workflow. That tool can then be installed as usual in a Galaxy server. It will allow the Taverna workflow to become part of a Galaxy pipeline. Users do not need to install the generator application themselves. It is available as part of the myExperiment (http://www.myexperiment.org) website. While browsing Taverna2 workflows the user will see an additional link "Download as Galaxy Tool" in the download section of the workflow. The next page provides instructions (and references) as well as an option to download the current workflow as a Galaxy tool. For more information and contact details: https://trac.nbic.nl/elabfactory/wiki/eGalaxy Cheers, Kostas

2 2

Use of galaxy with Illumina Pipeline RTA / OLB/ CASAVA / GERALD etc
by WATSON Mick 06 Apr '11

06 Apr '11

Hi I'd be interested to hear from anyone who has tried, or even thought about, using galaxy to manage the Illumina pipeline (RTA, OLB, CASAVA, GERALD and all those familiar names). Thanks Mick The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0

Relative file path in Galaxy
by Zhe Chen 06 Apr '11

06 Apr '11

Hi, My tool has a script reads a file in the same directory as the script. When I try to use relative path to read the file, it works by directly run it, but does not work when calling from Galaxy. Absolute path will work, but I want to know is there a way to do it using relative path. Can you give me some suggestion? myTool.pl use the following code the read the file in the same directory my $genusfiles = "genus.txt"; open (GENUSGP, "$genusfiles") or die "$genusfiles: $! \n"; close(GENUSGP); Thanks

2 1

SGE and Galaxy (a different approach)
by andrew stewart 06 Apr '11

06 Apr '11

I'm aware of how to configure Galaxy to use SGE in universe_wsgi.ini, however what I want to do is a little different. Because I only want certain processes to be submitted to the queue, I'd rather control this at the tool configuration level (the xml wrapper). For example: <command interpreter="bash"> qsub myscript.sh </command> This will work, except that the status of the job (in Galaxy) shows as completed even though the job has simply been submitted to SGE. Basically Galaxy 'loses track' of the process because the submission process (myscript.sh) has completed even if the actual job hasn't. Has anyone else tried anything like this before, or have anything helpful to suggest? One thought is to somehow cause the myscript.sh process to pause until the SGE job has completed... somehow. Any advice appreciated. Thanks, Andrew

6 7

Freebayes
by Juan Carlos Perin 05 Apr '11

05 Apr '11

I'm trying to run Freebayes SNP calling on a bam file from a BWA alignment of solid data. I keep getting the following error: [bam_sort_core] merging from 40 files... freebayes: invalid option -- Y did you mean --bam ? I'm not sure where this is set in the code to try to debug things. I'm wondering if anyones had this error, and how to fix it? Just to note, I'm running Brad Chapmans branch of galaxy for the LIMS functionality. But it appears to be an issue with either galaxy-dist or this one. Thanks in advance for any advice. Juan Perin

2 4