Hi Dannon,
Thank for the pointers from before, I have a version of galaxy that is able to change runtime parameters through the API when running a workflow.

Format: param=<tool name>=<tool parameter>=<value>

# example execution: (currently changing 2 parameters: 1) bowtie, suppressHeader parameter 2) Chip-seq peakcalling, aligner change
python workflow_execute.py api_key http://localhost:8080/api/workflows workflow_galaxy_id 'Test API' '69=ld=a799d38679e985db' '70=ld=33b43b4e7093c91f' 'param=peakcalling_spp=aligner=arachne' 'param=bowtie_wrapper=suppressHeader=True'

However, I was wondering if you had any pointers on how to deal w/ changing species or chromosome lengths?

# example workflow: bowtie tool
           "tool_id": "bowtie_wrapper",
           "tool_state": "{\"suppressHeader\": \"\\\"False\\\"\", \"refGenomeSource\": \"{\\\"genomeSource\\\": \\\"indexed\\\", \\\"index\\\": \\\"dm3\\\", \\\"__current_case__\\\": 0}\", \"__page__\": 0, \"chromInfo\": \"\\\"/data/home/galaxy/galaxy-dist/tool-data/shared/ucsc/chrom/dm3.len\\\"\", \"singlePaired\": \"{\\\"sInput1\\\": null, \\\"sParams\\\": {\\\"sMismatchSeed\\\": \\\"2\\\", \\\"sUnmappedFile\\\": \\\"False\\\", \\\"sTrimH\\\": \\\"0\\\", \\\"sTrimL\\\": \\\"0\\\", \\\"sSuppressAlign\\\": \\\"1\\\", \\\"sSettingsType\\\": \\\"full\\\", \\\"sSeed\\\": \\\"-1\\\", \\\"sMismatchQual\\\": \\\"70\\\", \\\"sMaqSoapAlign\\\": \\\"-1\\\", \\\"sAlignLimit\\\": \\\"-1\\\", \\\"sTryHard\\\": \\\"noTryHard\\\", \\\"sRounding\\\": \\\"round\\\", \\\"__current_case__\\\": 1, \\\"sSkip\\\": \\\"0\\\", \\\"sBestOption\\\": {\\\"snMaxBacktracks\\\": \\\"125\\\", \\\"sBest\\\": \\\"noBest\\\", \\\"__current_case__\\\": 0}, \\\"sAllValAligns\\\": \\\"noAllValAligns\\\", \\\"sOffrate\\\": \\\"-1\\\", \\\"sSeedLen\\\": \\\"28\\\", \\\"sValAlign\\\": \\\"1\\\", \\\"sMaxFile\\\": \\\"False\\\"}, \\\"sPaired\\\": \\\"single\\\", \\\"__current_case__\\\": 0}\"}",
           "tool_version": "1.1.2",
           "type": "tool",
           "user_outputs": []

Is there a way to configure bowtie in galaxy to use the species automatically associated with input file? I am currently having troubles dealing w/ parameters that are defined in sub dictionaries defined in the tool state. Any general approaches would be appreciated. 

Thanks!
Richard 



On Mon, Dec 5, 2011 at 6:32 PM, Dannon Baker <dannonbaker@me.com> wrote:
> Richard,
>
> You're correct in that currently the workflow API affords no method for runtime modification of tool parameters, other than inputs.  Depending on your needs, it might be feasible to have a few static workflows that you reuse often via the workflow API.  If that isn't the case, and you think it'd be likely that you'll want to, say, programmatically modify parameters at runtime based on input data heuristics, then the API does need to be extended.  The first place to look at would be lib/galaxy/web/controllers/workflow.py (run() method), to see the general approach to running workflows and providing parameters in the existing infrastructure.  Then, see the API version at lib/galaxy/web/api/workflows.py (create() method) for a comparison.
>
> Let me know if you need any help with this at all.  Once it's done, if you're willing, we'd certainly like to merge the changes back in and include the functionality in the base galaxy distribution.
>
> -Dannon
>
>
>
>
> On Dec 5, 2011, at 5:10 PM, Richard Park wrote:
>
>> Hello,
>> I was wondering what would be the best way to extend Galaxy's API
>> functionality to allow for runtime modification of tool parameters?
>>
>> I have successfully been able to run workflows programmatically using
>> the API, following the basic steps in:
>> scripts/api/execute_workflow.py.
>> scripts/api/example_watch_folder.py
>>
>> However, it is unclear to me, what would be the best way to run
>> workflows through API with specific parameters at various steps.
>> Should I generate new workflows for every workflow that requires
>> different parameters and upload this to galaxy? Or would it be better
>> to extend the API to allow for runtime parameter configuration? If
>> this is the case, any pointers on how to extend this would be greatly
>> appreciated.
>>
>> Thank you,
>> Richard Park
>> ___________________________________________________________
>> 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/
>