Re: [Myexperiment-discuss] Upload any type of Workflow you want...

[Coles S.J.]

QSAR would be an excellent test case for myExperiment, as would Will's 
OpenMicroscopy data.

I like to think of myExperiement as a place to share PROCESS information and 
data. It just so happens that Bioinformaticians workflows are the low hanging 
fruit in that there is already a community built up and they have something in 
a useable/understandable format that they want to share!! That is why they were 
chosen as the first step to enlightenment! There is more to come from different 
disciplines, but they are harder to master. At this point in time you are 
correct - myExpt is workflow centric, but I hope that it is in a transition 
phase! All research is performed as some kind of 'workflow' or other but it can 
be exceptionally difficult to capture digitally.

We are looking at its suitability for a variety of chemistry data / experiments 
(synthesis, analytical instruments, computational simulations) and QSAR would 
be a good intermediate step. This is in the planning stages right now and 
therefore it could be factored in as another use case. Containing files and 
describing them through tags etc is possible but the tricky part is describing 
the process and linking to inputs/outputs for each step in a machine 
readable/reproducible way.  We are looking at stringing together (and 
describing in a digitally exchangable way) processes that arise from humans 
doing things with physical stuff (i.e. Cookery recipies with chemicals) and 
analytical instruments producing data and simulations that occur in a 
computer...once you step outside the boundary of the digital world and start 
interfacing with the real world describing processes in a workflow/myExpt way 
becomes really hard and will require quite a lot of bespoke plugin 
development...UsefluChem's EXPLANS  will be

Simon.

Simon Coles
Manager, EPSRC National Crystallography Service
School of Chemistry
University of Southampton
Southampton, UK. SO17 1BJ.
t: +44(0)2380 596722
f: +44(0)2380 596723
e: address@hidden
w: http://www.ncs.chem.soton.ac.uk
--



On 23/10/2008 13:23, "Ola Spjuth" <address@hidden> wrote:

Hi,

I find Gernot's question very interesting and would like to enquire
about something similar. We are developing a new standard for
exchanging QSAR experiments in a new XML format. We were thinking of
setting up our own repository, but I'd much rather use myExperiment
for this. In a way you could call the setup of a QSAR experiment for a
workflow (is this the terminology you'd like us to use?).

Anyway, having QSAR experiments in myExperiment would require some
specific modules/plugins in myExperiment to render/present/summarize
the QSAR xml files for users, as well as means to validate them
against a schema etc. What is the procedure for plugging specific
components into myExperiment? Are there any docs for this? How would
you make them available for other users to install and use (will it be
like facebook apps)? I briefly studied the myExperiment wiki but found
no pointers to develop such plugins. Do we need to set up our own
myExperiment server in order to customize it to our needs, and in that
case is it possible to keep this synchronized with the main
myExperiment repository?

A general question: myExperiments seems currently very workflow-
centric. Judging from the name (myExperiment, not myWorkflow) I assume
that the site is aimed for extensions towards other types of
experiments (I guess this is a question of definitions, as everything
in principle can be called a workflow). Are there other extensions you
are working towards to broaden myExperiments, or do you envision
several myExperiment sites focused towards different domains?

Looking forward to learning more about this :-)

Cheers,

    /Ola Spjuth



On 23 okt 2008, at 08.44, Gernot Stocker wrote:

> Hi myExperiment -Team,
> congratulations for this great new feature and it brings me to the
> question whether it could
> be also used to exchange xml-based experiment protocols? Within the
> next days we are
> going to release a new version of our software iLAP which offers
> despite other features
> some kind of ELN functionality. There, experiment protocols can be
> exported as xml
> "workflows" which are the basis for a web based data aquisition
> wizard.
>
> First, is it allowed to exchange also iLAP protocols via
> myExperiment? Which format do you
> use for your workflows in order to achieve a better integration of
> our protocols within
> myExperiment? Or can we use our own? Is there a developer
> documentation available for
> such an integration path?
>
> Sorry for the newbe questions, but your announcement encouraged me
> to follow this idea
> with more energy.
> Thanks,
> GS
>
> On Thursday 23 October 2008, David De Roure wrote:
>> The myExperiment team is very pleased to announce that myExperiment
>> can now store any type of workflow, not just Taverna ones.  This
>> means that you can upload any file as a workflow which can then be
>> shared, discovered, downloaded, tagged, rated, commented on, added
>> to packs and so on.
>>
>> We already have some Trident workflows, and some experimental plans
>> from chemistry (Usefulchem EXPLANS). You can upload your own types
>> of workflows by going to the upload workflow page (you may be asked
>> to log in). If we can't automatically detect the type of workflow
>> and infer data from it (like title and description) you can provide
>> this by selecting the "Enter custom metadata" option in step 2.
>>
>> Note that we currently provide extended support for Taverna 1
>> workflows only - we can automatically detect the type, parse title/
>> description metadata, generate preview images, view the internals
>> and enact these workflows all directly in myExperiment. In the
>> coming months we aim to work with our user communties to provide
>> extended support for other types of workflows.
>>
>> If you would like to get involved in providing extended support for
>> other types of workflows then please contact us. We are keen to
>> work with you in getting myExperiment to support as many other
>> workflows types as possible.
>>
>> Thanks again for all your support!
>>
>> The myExperiment Team
>>
>> http://www.myexperiment.org/announcements/20
>>
> --
> Gernot Stocker,
> Institute for Genomics and Bioinformatics(IGB)
> Petersgasse 14, 8010 Graz, Austria
> Tel.: ++43 316 873 5345
> http://genome.tugraz.at
> _______________________________________________
> Myexperiment-discuss mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/myexperiment-discuss



_______________________________________________
Myexperiment-discuss mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/myexperiment-discuss