[kepler-dev] RE: [SDM-SPA] Latest SPA Release (updated)

[dcolonn@ncsu.edu]

Hey Ilkay,

Thank for the suggestion.  I can't send you the code, because my laptop
just went in for repairs. But I can descibe the details of what I'm doing,
and maybe you can get our generic actor working with this service.

A good place to start is to get the Soaplab client code from the anonymus CVS

cvs -d :pserver:anonymous at cvsebi.ebi.ac.uk:/ebi/cvs/public login
Password: none
cvs -d :pserver:anonymous at cvsebi.ebi.ac.uk:/ebi/cvs/public co soaplab
cvs -d :pserver:anonymous at cvsebi.ebi.ac.uk:/ebi/cvs/public logout

The program AnalysisClient.java lets you run tfscan.  An example set of
parameters that works:


java AnalysisClient
   -e http://industry.ebi.ac.uk/soap/soaplab
   -name nucleic_transcription::tfscan
   -w -r
   sequence_usa embl:xlrhodop


If you want to see all the inputs and outputs add -ii and -oo.

java AnalysisClient
   -e http://industry.ebi.ac.uk/soap/soaplab
   -name nucleic_transcription::tfscan
   -ii -oo

To see an xml description of parameters use -d.

There are 2 ways to input sequance data.  I want to start with this one

sequence_direct_data gccccccccgt

but you can also point to an EMBL database entry

sequence_usa embl:xlrhodop

As another approach, you can generate stubs from the WSDL document but you
need the latest (CVS build) version of apache axis wsdl2java to deal with
the <throwable> tag. So, if my ****ing laptop was working , my approach
would be to use the analysis client.


-Daniel







>
> Hi Dan,
>
> This sounds great.
>
> I think we can directly use it with the generic web service actor as it
> is available as a soap service. Developing a wrapper for it might be
> unnecessary.
>
> I will be happy to try it if you could provide me with the operation
> name and inputs for it.
>
> What I saw from wsdl is that this web service uses almost all soap
> types. Currently the ws actor handles basic xsd types and some array
> types. It should be possible to use it as is with the extension to the
> web service actor for the map array.
>
> Ilkay
>
>
>
> -----Original Message-----
> From: sdm-dev-admin at sdsc.edu [mailto:sdm-dev-admin at sdsc.edu]On Behalf Of
> dcolonn at ncsu.edu
> Sent: Monday, March 08, 2004 10:10 AM
> To: buttler1 at llnl.gov
> Cc: Daniel Colonnese; sdm-dev at sdsc.edu; sdm at renoir.csc.ncsu.edu;
> kepler-dev at ecoinformatics.org; xin2 at llnl.gov
> Subject: Re: [SDM-SPA] Latest SPA Release (updated)
>
>
>
> Soaplab is hosting a WSDL service for tfscan.  The SoapLab developer,
> Martin Senger, just updated the TRANSFAC data on his end.
>
> I was the first on to use this service, so we worked out a few minor
> problems; now it's a working web service.
>
> The tfscan WSDL is at:
>
> http://industry.ebi.ac.uk/soaplab/wsdl/nucleic_transcription__tfscan.wsdl
>
>
> Since you think it is a possible replacement for the old service,  I'll
> get to work on wrapping a ptolemy actor for this tfscan.
>
> And I'll keep my eyes open for a computer scientist who's fluent in
> German.
>
>
> -Dan
>
>
>> Hi Daniel,
>>
>> I believe that this would be a suitable replacement for Transfac.  My
>> understanding is that to use this tool, you need to download the data
>> directly from the transfac site.  Did you have a chance to check that
>> the transfac FTP site is still available?  When I tried, the response
>> just hung.  If we could set up the service locally (either as an
>> executable to include in a distribution, or as a Web service) it would
>> seem to have some value. Perhaps I misunderstood your intent though; did
>> you mean to say that they (the developers of EMBOSS) would be willing to
>> host a WSDL service for the world to access?
>>
>> Many of the services provided by the EMBOSS tool suite that includes
>> tfscan would make useful additions to the bioinformatics side of
>> Kepler.  Once we get the system set up here for Matt, we will explore
>> what would be of value to him.
>>
>> As a side note, it turns out that Matt Coleman has a seat license for
>> genomatix.de, which includes a transfac web interface.  So, it should be
>> possible to develop an actor that uses this information.  Perhaps one of
>> us could contact this company directly to see if they would be willing
>> to make their services available as a Web service for licensed users.
>> This should save us some implementation/screen scraping effort and
>> sounds like a reasonable approach for the company. Any volunteers?
>>
>> Two last pieces of information from my meeting with Matt today:
>> He is going to be on travel for most of March.  We will meet with him
>> again on the 26th to show him the workflow system.
>> Also, in the future, he would be interested in joining our monthly SPA
>> meeting, or perhaps other ad hoc meetings, to discuss the state of the
>> system and what could be done to best meet his needs.
>>
>> Dave
>>
>> Daniel Colonnese wrote:
>>
>>>Dear All,
>>>
>>>I have found a potential TRANSFAC service. It is a free service offered
>>>by European Bioinformatics Institute.  I have been in touch with the
>>>service author and they are cooperative and willing to work us.
>>>
>>>Before I turn this into a Ptolemy actor, I need some help to determine
>>>whether this service is suitable replacement for the previous transfac
>>>service.
>>>
>>>The service is an interface to the TFSCAN program.  Here is a
>>>description of that program
>>>
>>>http://www.hgmp.mrc.ac.uk/Software/EMBOSS/Apps/tfscan.html
>>>
>>>Let's schedule a meeting for Monday or Tuesday where Matt can help us
>>>decide.
>>>
>>>Please let me know,
>>>
>>>Dan
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>
>