Thanks Jim for giving this extensive update on how to improve the Qiime 
wrappers. 

We had contact about the Qiime wrappers before, and I have been using them in 
Galaxy on a weekly basis. Therefore I can say they have been proven to be very 
useful. I mainly edited the parameter and output sections of some configuration 
files, but all still use the wrapper script. The most important scripts are the 
split libraries and the workflow scripts (pick_otus_through_otu_table, 
alpha/beta diversity workflows).

I use the Qiime scripts in a Galaxy Cloudman environment adjusted to a 
multicore server. Setup on EC2 shouldn't be a problem, although you need to 
place the requirement tag for all the needed dependecies on top of the xml file.

I will have a look at the suggestions and I am willing to help to improve Qiime 
support for galaxy either by working on the configuration files or testing them 
out. So please keep me in the loop...

Greetings,

Mattias


-----Original Message-----
From: galaxy-dev-boun...@lists.bx.psu.edu on behalf of 
galaxy-dev-requ...@lists.bx.psu.edu
Sent: Sat 12/31/2011 6:00 PM
To: galaxy-dev@lists.bx.psu.edu
Subject: galaxy-dev Digest, Vol 66, Issue 26
 
Send galaxy-dev mailing list submissions to
        galaxy-dev@lists.bx.psu.edu

To subscribe or unsubscribe via the World Wide Web, visit
        http://lists.bx.psu.edu/listinfo/galaxy-dev
or, via email, send a message with subject or body 'help' to
        galaxy-dev-requ...@lists.bx.psu.edu

You can reach the person managing the list at
        galaxy-dev-ow...@lists.bx.psu.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of galaxy-dev digest..."


HEY!  This is important!  If you reply to a thread in a digest, please
1. Change the subject of your response from "Galaxy-dev Digest Vol ..." to the 
original subject for the thread.
2. Strip out everything else in the digest that is not part of the thread you 
are responding to.

Why?
1. This will keep the subject meaningful.  People will have some idea from the 
subject line if they should read it or not.
2. Not doing this greatly increases the number of emails that match search 
queries, but that aren't actually informative.

Today's Topics:

   1. Re: Existing efforts to convert the QIIME pipeline to Galaxy?
      (Jim Johnson)


----------------------------------------------------------------------

Message: 1
Date: Thu, 29 Dec 2011 14:02:04 -0600
From: Jim Johnson <johns...@umn.edu>
To: Jose Carlos Clemente <jose.cleme...@gmail.com>,
        "galaxy-dev@lists.bx.psu.edu" <galaxy-dev@lists.bx.psu.edu>
Cc: Greg Caporaso <gregcapor...@gmail.com>,     Rob Knight
        <rob.kni...@colorado.edu>,      Jesse Stombaugh <jistomba...@gmail.com>,
        Zuzolo Amanda <azuz...@gmu.edu>, Antonio Gonz?lez
        <antgo...@gmail.com>,   Gillevet Patrick <pgill...@gmu.edu>, 
j...@umn.edu,
        Daniel McDonald <was...@gmail.com>
Subject: Re: [galaxy-dev] Existing efforts to convert the QIIME
        pipeline to Galaxy?
Message-ID: <4efcc73c.7010...@umn.edu>
Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"


It is easiest to generate tools for galaxy when the applications or scripts can 
take arbitrarily named input files and generate output to given path names.
Input directories, output directories are very convenient on the command line, 
but more of a challenge when crafting a galaxy tool.
That said, many applications require a wrapper script to work with in galaxy.
Thank you for the consistent script_info[] help/usage syntax in the qiime 
scripts,  which enabled me to generate a skeleton galaxy tool_config file for 
each qiime script.

I had some time last spring to work on integrating qiime into galaxy.
Unfortunately, I haven't had any time since to work on this.
I put those partial results  on the Galaxy Tool Shed:    
http://toolshed.g2.bx.psu.edu/
There's a continuing effort at George Mason University to incorporate qiime 
into galaxy tools, so you may want to ask them what they need.


I started by generating galaxy tool_config files, e.g. align_seqs.xml,  by 
using python to get the script_info[] from the qiime script:

$ cat generate_tool_config.bash
#!/usr/bin/env bash
python $1 > ${1%.*}.help
cat tool_template.txt | sed "s/__TOOL_BINARY__/${1}/" | python -i $1 -h > 
${1%.*}.log

(I'll attach tool_template.txt )

This generated skeleton tool_config .xml files that I could then edit as needed.
( http://wiki.g2.bx.psu.edu/Admin/Tools/Tool%20Config%20Syntax )

I originally was calling all qiime scripts from a tool wrapper:  
qiime_wrapper.py
But, if a script can be called with any input filepaths and write its results 
to any filepaths, and only writes to STDERR when it fails, then you could call 
that script directly.


When should you use a tool_wrapper or call the qiime script directly?
   Many of the qiime scripts could probably be called directly, especially if 
it can be called with arbitary input/output file pathnames.
   The reasons for using a tool wrapper may be if input/output needs to be 
manipulated, moved, renamed in order to be used by the qiime script.
   You'll also need a tool wrapper if the names or number of the output files 
can not be determined from the parameter settings.
   ( http://wiki.g2.bx.psu.edu/Admin/Tools/Multiple%20Output%20Files )
   If your tool relies on a file ext to determine a format, you'll have to 
rename the input.
   ( Galaxy dataset pathnames will look something like:  
/<your_galaxy_file_path>/072/dataset_72931.dat )
   The format/type of a dataset is stored in its metadata, so the tool_config 
can use that information, especially if a script can take muliple alternative 
input formats.
   A tool_wrapper can also be used to manage the stdout or stderr from a tool.  
 Galaxy currently interprets any output on stderr as a failure.



A couple changes in galaxy should make somethings easier than when I first 
attempted this:
   - galaxy now accepts dataset requests with sub directories. ( 
https://bitbucket.org/galaxy/galaxy-central/issue/494/support-sub-dirs-in-extra_files_path-patch
 )
     That means that output HTML files with links into sub directories can be 
left intact, with the html copied to the output dataset and the linked files to 
its "extra_files_path".
   - if you know the pathname of an output relative to the working directory, 
galaxy can copy it automatically to the output dataset using the from_work_dir 
attribute.
     ( see example in:  
https://bitbucket.org/galaxy/galaxy-central/src/21b645303c02/tools/ngs_rna/tophat_wrapper.xml
 )

Datatypes
   You may want to create new datatypes to make it easier for the user to 
correctly select inputs to a tool from previous outputs.
   For example, the qiime mapping file is a tabular file with specific 
requirements.  I put a 'qiimemapping' datatype in 
lib/galaxy/datatypes/metagenomics.py and datatypes_conf.xml
   so an input could generate a select list containing only qiimemapping 
datasets rather than all tabular ones.

Generating a configfile
   You can generate configfiles in the galaxy tool_config .xml file.   The 
configfile is generated by the Cheetah interpreter just as the commandline is.
   see:  alpha_rarefaction.xml

The qiime_wrapper.py was patterned after the mothur_wrapper.py   with some of 
the same wrapper params to handle run time determined output (perhaps not 
needed):
   --galaxy_datasets
          a comma separated list of regex:output_dataset the wrapper searches 
the working_dir and copies the file that matches the regex to the outout dataset
          if the exact pathname is known, use the "from_work_dir" attribute 
instead
   --galaxy_datasetid
          would be an output dataset id that would be used to dynamically 
create additional new datasets at job termination
          ( http://wiki.g2.bx.psu.edu/Admin/Tools/Multiple%20Output%20Files 
"Number of Output datasets cannot be determined until tool run")
   --galaxy_new_datasets
          a comma separated list of regex:datatype used to dynamically create 
additional new datasets at job termination
   --galaxy_new_files_path
          the galaxy dir for dynamically generated output datasets




On 12/14/11 12:11 PM, Jose Carlos Clemente wrote:
> Hi Jim,
>
> we just had a meeting to discuss ideas for a QIIME GUI, and were wondering 
> how far did you manage to get with your plan to develop a QIIME wrapper for 
> Galaxy. Do you have something working that we could take a look at? Were 
> there any particular issues with the integration? We might have some 
> bandwidth to work on this over next year, but thought it would be good to 
> check with you first.
>
> Thanks,
> Jose
>
>
> On Mon, Feb 7, 2011 at 20:24, Rob Knight <rob.kni...@colorado.edu 
> <mailto:rob.kni...@colorado.edu>> wrote:
>
>     Hi Jim,
>
>     Just to echo Greg's comments; it would be excellent to connect QIIME to 
> Galaxy: let me know what is necessary from my end to ensure that this happens.
>
>     Thanks,
>     Rob
>
>     On Feb 7, 2011, at 12:30 PM, Greg Caporaso wrote:
>
>>     Hi Jim,
>>     This is great! We're very enthusiastic about hooking up QIIME with 
>> Galaxy, and definitely encourage you to work on it. It's something we've 
>> discussed doing, but no one in our lab is actively working on it. One 
>> particular use case that I've been thinking about would be to set up 
>> Galaxy/QIIME in our QIIME EC2 images so our EC2 users could access their 
>> images via the Galaxy web server. Are you planning to release all of the xml 
>> files under an open source/free distribution model?
>>
>>     I'm definitely interested in progress updates. Let me know if there are 
>> ways that we can help out, for example by giving you a list of what we 
>> believe are the most commonly used QIIME scripts so you can focus your 
>> efforts on those. I'm also cc'ing some of the folks in the lab here who are 
>> interested in Galaxy integration.
>>
>>     Greg
>>
>>     On Mon, Feb 7, 2011 at 10:36 AM, Jim Johnson <johns...@umn.edu 
>> <mailto:johns...@umn.edu>> wrote:
>>
>>         Greg,
>>
>>         Tim te Beek let me know that you have an interest in incorporating 
>> Qiime into Galaxy.   I'm currently working on that in support of a grant at 
>> the University of Minnesota.   I'll keep you apprised of my progress if you 
>> are interested.
>>
>>         Thanks,
>>
>>         JJ
>>
>>         James Johnson
>>         University of Minnesota Supercomputing Institute
>>         117 Pleasant St. SE
>>         Minneapolis, MN 55455
>>         Email: j...@umn.edu <mailto:j...@umn.edu>
>>
>>
>>
>>         -------- Original Message --------
>>         Subject:     Re: [galaxy-dev] Existing efforts to convert the QIIME 
>> pipeline to Galaxy?
>>         Date:        Mon, 07 Feb 2011 11:15:38 -0600
>>         From:        Jim Johnson <johns...@umn.edu> <mailto:johns...@umn.edu>
>>         Reply-To:    j...@umn.edu <mailto:j...@umn.edu>
>>         Organization:        MSI University of Minnesota
>>         To:  Tim te Beek <tim.te.b...@nbic.nl> <mailto:tim.te.b...@nbic.nl>
>>
>>
>>
>>         Tim,
>>
>>         Sorry to hear we won't be doing this together.   I'm finding some 
>> sporadic time to work on this.  I also put my Mothur metagenomics wrapper 
>> code on http://community.g2.bx.psu.edu/
>>
>>         I looked at the initial work you did, and combined that with the 
>> script design ideas I used with Mothur.   I'm writing a single script to 
>> call all the Qiime scripts, and generating a tool config (.xml) for each 
>> qiime script.   Since the qiime scripts all contain a rather consistent 
>> script_info dictionary in the code, I wrote a script to generate an initial  
>> tool config from that script_info.   I'm now starting the tedious process of 
>> editing those for each script.  I'll attach the few I've started on.  If you 
>> have any users that would able to beta test this when I have this more 
>> complete, please let me know.
>>
>>         Thanks,
>>
>>         JJ
>>
>>
>>         On 2/7/11 10:15 AM, Tim te Beek wrote:
>>>         Hi Jim,
>>>
>>>         Sorry for the lack of feedback / updates on my part in converting 
>>> QIIME to Galaxy. Following the initial difficulties with the conversion, 
>>> the support project was put on hold pending further discussion. This 
>>> morning the clients decided to forgo the Galaxy implementation, as their 
>>> implementation would only ever be used by a single (technical) user, 
>>> reducing the benefit of having a user friendly graphical interface.
>>>
>>>         So I'm sorry to say I won't be able to contribute to the 
>>> conversion, although your wrapper script did look like a viable solution to 
>>> work around the unknown output file problems. Should you be able to 
>>> complete the conversion yourself and be willing to share your 
>>> implementation, I would still very much like to know.
>>>
>>>         Perhaps the following could still be of some use to you: I 
>>> contacted the QIIME development group about adding output file name 
>>> arguments to the scripts, as can be seen here:
>>>         
>>> http://sourceforge.net/tracker/?func=detail&atid=1157167&aid=3164813&group_id=272178
>>>  
>>> <http://sourceforge.net/tracker/?func=detail&atid=1157167&aid=3164813&group_id=272178>
>>>         Although it appears they wont be able to add output file name 
>>> script parameters, they did say they're interested in Galaxy, and would be 
>>> willing to help if there's a need for that.
>>>
>>>         Additionally, we at NBIC do have a team that's more experienced at 
>>> using Galaxy and converting tools to work in Galaxy, and if you want I 
>>> could put you in contact with them about supporting this effort. In 
>>> informal discussions they did seem to look favorable upon your solution, so 
>>> perhaps that could be extended into a collaboration.
>>>
>>>         Best regards,
>>>         Tim
>>>
>>>         On Fri, Jan 14, 2011 at 20:01, Tim te Beek <tim.te.b...@nbic.nl 
>>> <mailto:tim.te.b...@nbic.nl>> wrote:
>>>
>>>             Ah ok, that's still fine. Thanks for the archive! I'll see what 
>>> I can learn from how you've handled running command and handling in- & 
>>> output for the non-trivial scripts, and apply that to the QIIME scripts.
>>>             Should you wish so I could get back to you if there's any 
>>> significant new developments on my part in moving the QIIME scripts to 
>>> Galaxy, other than the Trac / SVN changes.
>>>
>>>             Have a good weekend!
>>>             Tim
>>>
>>>
>>>             On Fri, Jan 14, 2011 at 19:12, Jim Johnson <johns...@umn.edu 
>>> <mailto:johns...@umn.edu>> wrote:
>>>
>>>                 I should probably be more clear, in that mothur and qiime 
>>> overlap in functionality and in the algorithms they use.
>>>                 They do use different data formats.
>>>
>>>
>>>
>>>                 On 1/14/11 12:02 PM, Tim te Beek wrote:
>>>>                 Hi Jim,
>>>>
>>>>                 If you could send me the Mothur package, that would be 
>>>> great. If there's indeed as much overlap then it could be a great way to 
>>>> kick-start the QIIME pipeline conversion development. I can look into it 
>>>> starting monday, with hopefully nothing else to distract me the whole week.
>>>>
>>>>                 Best regards,
>>>>                 Tim
>>>>
>>>>
>>>>                 On Fri, Jan 14, 2011 at 18:49, Jim Johnson 
>>>> <johns...@umn.edu <mailto:johns...@umn.edu>> wrote:
>>>>
>>>>                     Tim,
>>>>
>>>>                     I just started looking at qiime myself.   We are 
>>>> hoping to release a galaxy server to the University of Minnesota 
>>>> researchers in the next month,
>>>>                     so there has been much to do in preparation.   I'll 
>>>> take a look at what you've done.  Thanks.
>>>>
>>>>                     I finished a galaxy interface to the Mothur 
>>>> metagenomics package http://www.mothur.org/wiki/Main_Page.   It appears to 
>>>> have a lot of overlap with Qiime.
>>>>                     I generated a number of datatypes,  and made use of 
>>>> many of the  tool config concepts that are available in galaxy.
>>>>                     If you are interested,  I can send that to you:   
>>>> mothur.tar.gz  (55108 bytes).   I hope to submit it 
>>>> tohttp://community.g2.bx.psu.edu/ in the next month.
>>>>
>>>>                     Regards,
>>>>
>>>>                     JJ
>>>>
>>>>
>>>>                     On 1/14/11 9:52 AM, Tim te Beek wrote:
>>>>>                     Hi Jim,
>>>>>
>>>>>                     Sorry to only respond just now: I've been tied up in 
>>>>> other projects, so have not been able to get any work done on this up to 
>>>>> today.
>>>>>
>>>>>                     What I currently have is not much, since my progress 
>>>>> has been slowed by my unfamiliarity with both Galaxy and the QIIME tools 
>>>>> (I'm working on this as an outsider for a small support project). But 
>>>>> what I do have I've made available through the following Trac       / SVN 
>>>>> instance:
>>>>>                     _
>>>>>                     _
>>>>>
>>>>>                         
>>>>> https://trac.nbic.nl/brs2010p26/browser/trunk/galaxy_dist/tools/qiime
>>>>>
>>>>>
>>>>>                     These configuration files so far can run:
>>>>>
>>>>>                       * check_id_map
>>>>>                       * split_libraries
>>>>>                       * pick_otus_through_otu_table
>>>>>
>>>>>                     For the first two the output is correctly directed to 
>>>>> the Galaxy history, the last one generates quite a large set of output 
>>>>> files which might have to be made available otherwise (the documentation 
>>>>> on this gives a few options: 
>>>>> https://bitbucket.org/galaxy/galaxy-central/wiki/ToolsMultipleOutput)).
>>>>>
>>>>>                     I had some trouble working around the fact that most 
>>>>> QIIME scripts do not allow one to specify output filenames on the command 
>>>>> line, which seems to be a requirement for Galaxy to import files into the 
>>>>> history pane. What I've done is written a bash file to accompany each XML 
>>>>> file, that expects as first command line argument the command that should 
>>>>> be evaluated, and as additional arguments the output filenames expected 
>>>>> by Galaxy. After running a tool I then move whatever output files were 
>>>>> created to the paths expected by Galaxy. I must say I'm not (at all) sure 
>>>>> if this current approach is something to recommend, or if better 
>>>>> alternatives are available. Any feedback on this would be highly 
>>>>> appreciated.
>>>>>                     An additional source of pain was that the 
>>>>> split_libraries script requires specific file extensions for it's fasta & 
>>>>> quality files, which required passing them separately and symbolic 
>>>>> linking them as .fna & .qual files.
>>>>>
>>>>>                     So far I've not taken the time to introduce new QIIME 
>>>>> specific data types, but this could be something to consider given the 
>>>>> large amount of 'tabular' input and output files.
>>>>>
>>>>>                     I hope this email and the scripts linked above can 
>>>>> still be of some use to you (or vice-versa), but in either case any 
>>>>> feedback or collaboration would be much appreciated.
>>>>>
>>>>>                     Best regards,
>>>>>                     Tim
>>>>>
>>>>>
>>>>>                     On Wed, Nov 24, 2010 at 15:48, Jim Johnson 
>>>>> <ctof...@comcast.net <mailto:ctof...@comcast.net>> wrote:
>>>>>                     > Tim,
>>>>>                     >
>>>>>                     > I hope to also look at incorporating QIIME into our 
>>>>> local Galaxy instance at
>>>>>                     > the University of Minnesota, but probably won't be 
>>>>> able to start for a
>>>>>                     > couple weeks.  It would be good to develop that in 
>>>>> coordination with others.
>>>>>                     >
>>>>>                     > I just finished incorporating "Mothur" metagenomics 
>>>>> suite
>>>>>                     > http://www.mothur.org/ (Dr. Patrick Schloss,  
>>>>> Department of Microbiology &
>>>>>                     > Immunology at The University of Michigan) into our 
>>>>> Galaxy server at the
>>>>>                     > University of Minnesota.   I hope to contribute 
>>>>> that to
>>>>>                     > http://community.g2.bx.psu.edu/ after some testing 
>>>>> by our researchers.  If
>>>>>                     > the Galaxy wrappings for Mothur are of any interest 
>>>>> to you, I can send you a
>>>>>                     > copy any time.
>>>>>                     >
>>>>>                     > Thanks,
>>>>>                     >
>>>>>                     > JJ
>>>>>                     >
>>>>>                     > James E Johnson
>>>>>                     > Minnesota Supercomputing Institute
>>>>>                     > University of Minnesota
>>>>>                     >
>>>>>                     >
>>>>>                     > On Nov 23, 2010 at 07:22 AM, Tim te Beek wrote:
>>>>>                     >
>>>>>                     >> Hello all,
>>>>>                     >>
>>>>>                     >> Is anyone aware of any existing efforts to port 
>>>>> the QIIME sequencing
>>>>>                     >> pipeline (http://qiime.sourceforge.net/) to 
>>>>> Galaxy? I would like to run
>>>>>                     >> QIIME analyses through Galaxy to get better 
>>>>> control of
>>>>>                     >> intermediate processing steps, but before I start 
>>>>> to convert (a subset of)
>>>>>                     >> some 90 scripts, I'd first like to make sure this 
>>>>> has not been done before
>>>>>                     >> by anyone willing to share their work.
>>>>>                     >>
>>>>>                     >> So: has anyone converted the QIIME pipeline to 
>>>>> Galaxy before, and would
>>>>>                     >> they
>>>>>                     >> be willing to share their scripts?
>>>>>                     >>
>>>>>                     >> Best regards,
>>>>>                     >> Tim
>>>>>                     >
>>>>>                     > _______________________________________________
>>>>>                     > galaxy-dev mailing list
>>>>>                     > galaxy-dev@lists.bx.psu.edu 
>>>>> <mailto:galaxy-dev@lists.bx.psu.edu>
>>>>>                     > http://lists.bx.psu.edu/listinfo/galaxy-dev
>>>>>                     >
>>>>>
>>>>
>>>
>>>
>>>
>>
>>
>>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
<http://lists.bx.psu.edu/pipermail/galaxy-dev/attachments/20111229/286f3564/attachment-0001.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: tool_template.txt
URL: 
<http://lists.bx.psu.edu/pipermail/galaxy-dev/attachments/20111229/286f3564/attachment-0001.txt>

------------------------------

_______________________________________________
galaxy-dev mailing list
galaxy-dev@lists.bx.psu.edu
http://lists.bx.psu.edu/listinfo/galaxy-dev


End of galaxy-dev Digest, Vol 66, Issue 26
******************************************

<<winmail.dat>>

___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/

Reply via email to