sed-ml-discuss — Migrating to google groups: https://groups.google.com/d/forum/sed-ml-discuss

Re: [SED-ML-discuss] XPath explanations in the spec

[David N. <dav...@gm...>]

> Surely the prefix -> namespace URI binding should be contained within the
> SED-ML file?  That's what happens in XSLT after all.  So somewhere in the
> SED-ML document (e.g. on the root element) you'd use xmlns:prefix="url" to
> declare all the prefixes you are going to use in XPath expressions later in
> the document.

yep - I agree with Jonathan. and XSLT is a good example to follow...

Cheers,
David.

>
> Best wishes,
> Jonathan
>
> On 19/11/2010 20:41, Richard Adams wrote:
>
> On 19 Nov 2010, at 18:09, Frank T. Bergmann wrote:
>
>
> b) Ideally, XPath prefix = xmlns prefix in model XML. If empty prefix is
> used in XML, then there are several alternative options:
> i) XPath prefix should be a string contained in the Namespace URI declared
> in the model XML file.
>
> I'm for this one! we should not limit the namespaces to be used!
>
> Its easy to implement but there is at least the possibility of non-unique
> obvious prefixes.
> E.g., for a typical SBML Biomodel with the following namespace declarations:
> http://purl.org/dc/terms/
> http://purl.org/dc/elements/1.1/
> http://www.w3.org/1999/xhtml
> http://www.w3.org/1999/02/22-rdf-syntax-ns#
> http://biomodels.net/model-qualifiers/
> http://www.sbml.org/sbml/level2
> http://www.w3.org/1998/Math/MathML
> http://www.w3.org/2001/vcard-rdf/3.0#
> http://biomodels.net/biology-qualifiers/
> If you used the prefix 'purl' or 'dc' in XPath   for  elements (unlikely as
> it is you want to define  a change the model  annotation using a SEDML file)
> then this would not resolve unambiguously to a single namespace URI in the
> model.
>
> What I meant, was that instead of hard coding uri's or prefixes, we simply
> use whatever is defined in the model file. At the time of writing the xpath
> expression, the software tools have all information (and namespace sets)
> belonging to each XML node. So it really should be possible to generate an
> xpath expression to it.
>
> OK, so just to clarify -  for example, looking at some of the level 3
> extensions, where one might get several SBML-related namespaces in a single
> document - e.g., for the qualitative model package:
>
> xmlns="http://www.sbml.org/sbml/level2/version3" level="2" version="3"
>
> xmlns:qual="http://www.sbml.org/sbml/extension/qual/version1"
> then, when generating an XPath expression during SEDML creation,  it would
> be up to the software involved to work out to use the 'qual' prefix to
> identify qualitative sub-elements, and 'sbml' for any core sbml.  But to
> keep XPath prefixes identical to xmlns prefixes would be just a convention;
> it would need to be specified as a convention somwhere. Else, software could
> use qual_version1, or sbml_qual, or whatever as an XPath prefix. But unless
> we have some sort of convention, then the decoding of that XPath  prefix by
> a SEDML consumer is not guaranteed.
> And, the problem is worse with nested XML namespaces using the empty prefix,
> then there is no hint from the xmlns declaration as to what string the
> software should use as the XPath prefix.
> But perhaps I'm missing something here too.
> Cheers
> Richard
>
> Or perhaps I don't see the issue :)
> best
> Frank
> ------------------------------------------------------------------------------
> Beautiful is writing same markup. Internet Explorer 9 supports
> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
> Spend less time writing and  rewriting code and more time creating great
> experiences on the web. Be a part of the beta today
> http://p.sf.net/sfu/msIE9-sfdev2dev_______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
> Dr Richard Adams
> Software Development Team Leader,
> Centre For Systems Biology Edinburgh
> University of Edinburgh
> Tel: 0131 651 9019
> email : ric...@ed...
> Web: http://csbe.bio.ed.ac.uk/adams.php
>
>
>
>
>
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
> ------------------------------------------------------------------------------
> Beautiful is writing same markup. Internet Explorer 9 supports
> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
> Spend less time writing and  rewriting code and more time creating great
> experiences on the web. Be a part of the beta today
> http://p.sf.net/sfu/msIE9-sfdev2dev
>
> _______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
> ------------------------------------------------------------------------------
> Beautiful is writing same markup. Internet Explorer 9 supports
> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
> Spend less time writing and  rewriting code and more time creating great
> experiences on the web. Be a part of the beta today
> http://p.sf.net/sfu/msIE9-sfdev2dev
> _______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
>

Re: [SED-ML-discuss] XPath explanations in the spec

[Jonathan C. <jon...@co...>]

Hi all,

Surely the prefix -> namespace URI binding should be contained within 
the SED-ML file?  That's what happens in XSLT after all.  So somewhere 
in the SED-ML document (e.g. on the root element) you'd use 
xmlns:prefix="url" to declare all the prefixes you are going to use in 
XPath expressions later in the document.

Best wishes,
Jonathan

On 19/11/2010 20:41, Richard Adams wrote:
>
> On 19 Nov 2010, at 18:09, Frank T. Bergmann wrote:
>
>>
>>>>
>>>>> b) Ideally, XPath prefix = xmlns prefix in model XML. If empty 
>>>>> prefix is used in XML, then there are several alternative options:
>>>>> i) XPath prefix should be a string contained in the Namespace URI 
>>>>> declared in the model XML file.
>>>>
>>>> I'm for this one! we should not limit the namespaces to be used!
>>>
>>> Its easy to implement but there is at least the possibility of 
>>> non-unique obvious prefixes.
>>> E.g., for a typical SBML Biomodel with the following namespace 
>>> declarations:
>>>
>>> http://purl.org/dc/terms/
>>> http://purl.org/dc/elements/1.1/
>>> http://www.w3.org/1999/xhtml
>>> http://www.w3.org/1999/02/22-rdf-syntax-ns#
>>> http://biomodels.net/model-qualifiers/
>>> http://www.sbml.org/sbml/level2
>>> http://www.w3.org/1998/Math/MathML
>>> http://www.w3.org/2001/vcard-rdf/3.0#
>>> http://biomodels.net/biology-qualifiers/
>>>
>>> If you used the prefix 'purl' or 'dc' in XPath   for  elements 
>>> (unlikely as it is you want to define  a change the model 
>>>  annotation using a SEDML file) then this would not resolve 
>>> unambiguously to a single namespace URI in the model.
>>
>> What I meant, was that instead of hard coding uri's or prefixes, we 
>> simply use whatever is defined in the model file. At the time of 
>> writing the xpath expression, the software tools have all information 
>> (and namespace sets) belonging to each XML node. So it really should 
>> be possible to generate an xpath expression to it.
>
> OK, so just to clarify -  for example, looking at some of the level 3 
> extensions, where one might get several SBML-related namespaces in a 
> single document - e.g., for the qualitative model package:
>
> xmlns="http://www.sbml.org/sbml/level2/version3" level="2" version="3"
>
>     xmlns:qual="http://www.sbml.org/sbml/extension/qual/version1"
>
> then, when generating an XPath expression during SEDML creation,  it 
> would be up to the software involved to work out to use the 'qual' 
> prefix to identify qualitative sub-elements, and 'sbml' for any core 
> sbml.  But to keep XPath prefixes identical to xmlns prefixes would be 
> just a convention; it would need to be specified as a convention 
> somwhere. Else, software could use qual_version1, or sbml_qual, or 
> whatever as an XPath prefix. But unless we have some sort of 
> convention, then the decoding of that XPath  prefix by a SEDML 
> consumer is not guaranteed.
>
> And, the problem is worse with nested XML namespaces using the empty 
> prefix, then there is no hint from the xmlns declaration as to what 
> string the software should use as the XPath prefix.
>
> But perhaps I'm missing something here too.
>
> Cheers
> Richard
>>
>> Or perhaps I don't see the issue :)
>>
>> best
>> Frank
>>
>> ------------------------------------------------------------------------------
>> Beautiful is writing same markup. Internet Explorer 9 supports
>> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
>> Spend less time writing and  rewriting code and more time creating great
>> experiences on the web. Be a part of the beta today
>> http://p.sf.net/sfu/msIE9-sfdev2dev_______________________________________________
>> SED-ML-discuss mailing list
>> SED...@li...
>> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
> Dr Richard Adams
> Software Development Team Leader,
> Centre For Systems Biology Edinburgh
> University of Edinburgh
> Tel: 0131 651 9019
> email : ric...@ed... <mailto:ric...@ed...>
> Web: http://csbe.bio.ed.ac.uk/adams.php
>
>
>
>
>
>
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
>
> ------------------------------------------------------------------------------
> Beautiful is writing same markup. Internet Explorer 9 supports
> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2&  L3.
> Spend less time writing and  rewriting code and more time creating great
> experiences on the web. Be a part of the beta today
> http://p.sf.net/sfu/msIE9-sfdev2dev
>
>
> _______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss

Re: [SED-ML-discuss] XPath explanations in the spec

[Richard A. <ric...@ed...>]

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

Re: [SED-ML-discuss] XPath explanations in the spec

[Frank T. B. <fbe...@ca...>]

>> 
>>> b) Ideally, XPath prefix = xmlns prefix in model XML. If empty prefix is used in XML, then there are several alternative options:
>>>     i) XPath prefix should be a string contained in the Namespace URI declared in the model XML file.
>> 
>> I'm for this one! we should not limit the namespaces to be used!
> 
> Its easy to implement but there is at least the possibility of non-unique obvious prefixes.
> E.g., for a typical SBML Biomodel with the following namespace declarations:
> 
> http://purl.org/dc/terms/
> http://purl.org/dc/elements/1.1/
> http://www.w3.org/1999/xhtml
> http://www.w3.org/1999/02/22-rdf-syntax-ns#
> http://biomodels.net/model-qualifiers/
> http://www.sbml.org/sbml/level2
> http://www.w3.org/1998/Math/MathML
> http://www.w3.org/2001/vcard-rdf/3.0#
> http://biomodels.net/biology-qualifiers/
> 
> If you used the prefix 'purl' or 'dc' in XPath   for  elements (unlikely as it is you want to define  a change the model  annotation using a SEDML file) then this would not resolve unambiguously to a single namespace URI in the model.

What I meant, was that instead of hard coding uri's or prefixes, we simply use whatever is defined in the model file. At the time of writing the xpath expression, the software tools have all information (and namespace sets) belonging to each XML node. So it really should be possible to generate an xpath expression to it. 

Or perhaps I don't see the issue :) 

best
Frank

Re: [SED-ML-discuss] XPath explanations in the spec

[Richard A. <ric...@ed...>]

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

Re: [SED-ML-discuss] XPath explanations in the spec

[Frank T. B. <fbe...@ca...>]

Hello Richard, 

On Nov 19, 2010, at 8:42 AM, Richard Adams wrote:

> Dear All,
>  Just going through the SEDML spec again and I'm thinking we're a little light on description of XPath usage.

I'd be happy to see you add some more explanation. 

> I'd like to propose that we add a little explanation of valid XPath expressions to help with mapping namespaces to the XML document. There are several issues:
> 
> 1) At the least add some references to XPath tutorials etc.,

I'm not sure that references to XPath tutorials belong in the SED-ML specification, otherwise there should be links to XML tutorials, MathML tutorials and more 

> 
> 2) How to write an XPath query that will refer to the model elements correctly. Most of the example XPaths given as example in the spec are unprefixed - these will typically not work for most XML models which declare a namespace (e.g., 'xmlns="http://www.sbml.org/sbml/level2") but only with XML documents that declare no namespaces.

That is an oversight, as you can see in the threads that have come on this list before, we definitely need the prefix! Thank you for noting this. 

> 
> 3) In most SBML models in Biomodels,  both MathML and SBML elements use the empty prefix. However XPath expressions must provide prefixes to allow XPath to reconcile namespaces. E.g., if we want to refer to some Mathml we cannot write:
> //:reaction[@id='Reaction6']/:kineticLaw/:math
> as both MathML and SBML namespaces are using the empty prefix ""- XPath can't wok out which URI to use for the given prefix
> 
> So  propose that we add some guidelines. E.g.,
> a) All XPath element and attribute locators MUST be prefixed in SED-ML.

Yes that is it! I thought we had stated that in the specification, at least it was discussed to that end here. 

> b) Ideally, XPath prefix = xmlns prefix in model XML. If empty prefix is used in XML, then there are several alternative options:
>     i) XPath prefix should be a string contained in the Namespace URI declared in the model XML file.

I'm for this one! we should not limit the namespaces to be used!

>     ii) Use a set of recommended prefixes for all namespaces contained in supported models (e.g.., MathML->math, SBML->sbml,) This way we can ensure  unique prefixes at the expense of complete ignorance of the modelling language
>    iii) Require that all models used with SEDML do not use the empty prefix, and XPath prefix = model xmlns prefix (probably impractical).
>     
> 
> E.g., 
> //sbml:reaction[@id='Reaction6']/sbml:kineticLaw/math:math/math:apply
> 
> 
> This should allow tools to map XPath prefixes to URIs in the model with minimal manual intervention. I don't think we want to be in a position where tools need to provide UI for an enduser to resolve namespace issues before applying XPath changes -> so we need guidelines on creating them in a uniform way in the first place.
> 
> This is also in the tracker 3112732
> 
> Any better suggestions greatly appreciated...

Thanks for bringing this up again, 

best
Frank


> 
> Cheers
> Richard
> 
> 
> 
> Dr Richard Adams
> Software Development Team Leader,
> Centre For Systems Biology Edinburgh
> University of Edinburgh 
> Tel: 0131 651 9019
> email : ric...@ed...
> Web: http://csbe.bio.ed.ac.uk/adams.php
> 
> 
> 
> 
> 
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
> ------------------------------------------------------------------------------
> Beautiful is writing same markup. Internet Explorer 9 supports
> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
> Spend less time writing and  rewriting code and more time creating great
> experiences on the web. Be a part of the beta today
> http://p.sf.net/sfu/msIE9-sfdev2dev_______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss

[SED-ML-discuss] XPath explanations in the spec

[Richard A. <ric...@ed...>]

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

[SED-ML-discuss] SED-ML L1V1 Release Candidate 1

[Dagmar W. <dag...@un...>]

Dear all,


  we are happy to inform you that the SED-ML Level 1 Version 1 Release  Candidate 1 is now available for download from the sourceforge web site:
  https://sourceforge.net/projects/sed-ml

  Please feel free to comment on the specification and schema documents byadding items to the tracker on
  https://sourceforge.net/tracker/?group_id=293618


  Thanks to all of you for contributing to the discussions.
  Dagmar

Re: [SED-ML-discuss] id attribute for SEDML class?

[Dagmar W. <dag...@un...>]

Maybe it is time to be consistent and have IDs on ALL SED-ML elements.


On 09/23/2010 03:41 PM, Richard Adams wrote:
> Dear  SED-MLers,
>    I was wondering what anybody thought about adding some sort of
> identifier attribute to the top-level SEDML element? I think there
> might be a need for it, because at present there is no way of reliably
> distinguishing, programmatically, between different SEDML objects. Any
> tool or database  that could need to work with multiple SED-ML
> descriptions lacks a  business key to refer to these objects.
>
> In SBML, for example, this is covered by the  enclosing 'model'
> element with its 'id' attribute.
>
>
>    Cheers
>       Richard
>
>
>
>    

[SED-ML-discuss] SED-ML as workflow

[Fedor K. <fko...@gm...>]

Dear colleagues,

thinking about support of SED-ML in BioUML I recognized that it will be
quite natural
to present SED-ML document as a workflow and I think BioUML team will try
to
implement some spike solution about it.

What do you think about possibility to slightly redesign SED-ML to follow
the workflow
concept more explicitly: define a set of tasks and their inputs and outputs?

Additional benefit of this approach will be extencibility of SED-ML: next
versions
will simply define new tasks (for example, parameters optimisation).

With best regards,
Fedor

[SED-ML-discuss] id attribute for SEDML class?

[Richard A. <ra...@st...>]

Dear  SED-MLers,
  I was wondering what anybody thought about adding some sort of  
identifier attribute to the top-level SEDML element? I think there  
might be a need for it, because at present there is no way of reliably  
distinguishing, programmatically, between different SEDML objects. Any  
tool or database  that could need to work with multiple SED-ML  
descriptions lacks a  business key to refer to these objects.

In SBML, for example, this is covered by the  enclosing 'model'  
element with its 'id' attribute.


  Cheers
     Richard



-- 
Dr Richard Adams
Software Development Team Leader,
Centre For Systems Biology Edinburgh
University of Edinburgh
Tel: 0131 650 8285
email : ric...@ed...

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

[SED-ML-discuss] XML schema updates

[Dagmar W. <dag...@un...>]

Hello everyone,

there have been some smaller XML schema updates, also reflected in the 
current UML Schema and draft specification:

- the element "parameter" has an updated attribute "value" which is now 
of type "xs:double" (instead of "xs:string").
- the attributes "newXML" (in the changeXML element) and "math" (in the 
computeChange element) have been changed into elements.
- the "target" attribute in the change classes is now mandatory 
(required), in accordance with the UML schema.
- the logX, logY, logZ attributes of the curve and surface elements are 
now reguired. the default value "false" has been removed.


Let me know if you have any problems.
Best,
Dagmar

Re: [SED-ML-discuss] log axes on plots or lines

[Nicolas Le n. <le...@eb...>]

On 05/08/10 08:24, Dagmar Waltemath wrote:

> I vote for adding a default value for the logX/Y/Z attribute, setting it
> to "false" (pretty sure we have discussed that already!?).
> Checked in as
> http://sed-ml.svn.sourceforge.net/viewvc/sed-ml?diff_format=h&view=revision&revision=176
>
> Objections, please let me know.

I do. We should avoid any default values. SBML is moving away from them, 
and this is a good move. People tend to assume "common sense" 
defaults,without reading the specs. Problem is that in many cases "common 
sense" is not that common. Therefore I think the attribute should not in 
fact be optional.

-- 
Nicolas LE NOVERE, Computational Neurobiology, EMBL-EBI, Wellcome-Trust
Genome Campus, Hinxton CB101SD UK, Mob:+447833147074, Tel:+441223494521
Fax:468,Skype:n.lenovere,AIM:nlenovere,MSN:nle...@ho...(NOT email)
http://www.ebi.ac.uk/~lenov/, http://www.ebi.ac.uk/compneur/, @lenovere

Re: [SED-ML-discuss] log axes on plots or lines

[Dagmar W. <dag...@un...>]

Hej all,

just to summarise:

the log attribute is attached to the curve/surface elements.

The XML schema on SF was correct.
http://sed-ml.svn.sourceforge.net/viewvc/sed-ml?diff_format=h&view=revision&revision=174

The UML in the spec, and the overview tables were correct as well.
I have updated the textual description of the Plot2D/3D and 
Curve/Surface elements to comply with the UML.
Checked in on 
http://sed-ml.svn.sourceforge.net/viewvc/sed-ml?diff_format=h&view=revision&revision=177

I vote for adding a default value for the logX/Y/Z attribute, setting it 
to "false" (pretty sure we have discussed that already!?).
Checked in as 
http://sed-ml.svn.sourceforge.net/viewvc/sed-ml?diff_format=h&view=revision&revision=176

Objections, please let me know.

Sample files still have to be fixed.

Best,
Dagmar


On 08/04/2010 07:05 PM, Richard Adams wrote:
> Hello SEDMLers
>
>    There seems to be some discrepancies as to where the logX/logY/logZ
> attributes should go, from looking in the spec - E.g., section 4.6.2
> text says these attributes go in the Plot2d/3d elements, but the UML
> diagrams show these attributes in the Curve/Surface elements. The
> example files also differ, SEDML3.xml has them in 'Plot' but SEDML12
> and 21 have them in 'Curve'.
>    Which is correct?
>    Cheers
>      Richard
>
>
>
>    

Re: [SED-ML-discuss] log axes on plots or lines

[Nicolas Le n. <le...@eb...>]

On 04/08/10 18:57, Frank Bergmann wrote:

> unless someone shows me a plot that contains (in the same coordinate
> system) both logarithmic curves / segments and non-logarithmic ones I
> say it should be on the Plot2d/3d!.

This discussion already happened, and you already asked for examples, and 
we decided to put those attributes on curves and surfaces (they were 
initially, in Okinawa, on plots). Examples would be for instance energies 
put on the same figure than equilibrium constants. It happens very 
frequently. Admittedly, all the examples I saw used the hybrid log/linear 
only on the y axis of a 2D plot. But if we do it, let's do it in a general 
manner.

> Again this is not a transformation, it is a mode of representing the
> data. If a user would like the data to be transformed this should happen
> on the DataGenerators.

Exactly, and you may want to output a data in a linear way in a report and 
a log way in a curve for instance. A very practical example is BioModels DB 
models 1/2. The "useful" output, the physiologically relevant one, is the 
EPSC, that is the current in function of linear time, but the author 
presented log plots because the states transitions had very different 
time-scales.

> Cheers Frank
>
> On Aug 4, 2010, at 1:05 PM, Richard Adams wrote:
>
>>
>> Hello SEDMLers
>>
>> There seems to be some discrepancies as to where the logX/logY/logZ
>> attributes should go, from looking in the spec - E.g., section 4.6.2
>> text says these attributes go in the Plot2d/3d elements, but the UML
>> diagrams show these attributes in the Curve/Surface elements. The
>> example files also differ, SEDML3.xml has them in 'Plot' but SEDML12
>> and 21 have them in 'Curve'. Which is correct? Cheers Richard
>>
>>
>>
>> -- Dr Richard Adams Software Development Team Leader, Centre For
>> Systems Biology Edinburgh University of Edinburgh Tel: 0131 650 8285
>> email : ric...@ed...
>>
>> -- The University of Edinburgh is a charitable body, registered in
>> Scotland, with registration number SC005336.
>>
>>
>>
>> ------------------------------------------------------------------------------
>>
>>
The Palm PDK Hot Apps Program offers developers who use the
>> Plug-In Development Kit to bring their C/C++ apps to Palm for a share
>> of $1 Million in cash or HP Products. Visit us here for more details:
>> http://p.sf.net/sfu/dev2dev-palm
>> _______________________________________________ SED-ML-discuss mailing
>> list SED...@li...
>> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
>
> ------------------------------------------------------------------------------
>
>
The Palm PDK Hot Apps Program offers developers who use the
> Plug-In Development Kit to bring their C/C++ apps to Palm for a share of
> $1 Million in cash or HP Products. Visit us here for more details:
> http://p.sf.net/sfu/dev2dev-palm
> _______________________________________________ SED-ML-discuss mailing
> list SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss


-- 
Nicolas LE NOVERE, Computational Neurobiology, EMBL-EBI, Wellcome-Trust
Genome Campus, Hinxton CB101SD UK, Mob:+447833147074, Tel:+441223494521
Fax:468,Skype:n.lenovere,AIM:nlenovere,MSN:nle...@ho...(NOT email)
http://www.ebi.ac.uk/~lenov/, http://www.ebi.ac.uk/compneur/, @lenovere

Re: [SED-ML-discuss] log axes on plots or lines

[Frank B. <fbergman@u.washington.edu>]

Hej,

unless someone shows me a plot that contains (in the same coordinate system) both logarithmic curves / segments and non-logarithmic ones I say it should be on the Plot2d/3d!. 

Again this is not a transformation, it is a mode of representing the data. If a user would like the data to be transformed this should happen on the DataGenerators. 

Cheers
Frank

On Aug 4, 2010, at 1:05 PM, Richard Adams wrote:

> 
> Hello SEDMLers
> 
>  There seems to be some discrepancies as to where the logX/logY/logZ  
> attributes should go, from looking in the spec - E.g., section 4.6.2  
> text says these attributes go in the Plot2d/3d elements, but the UML  
> diagrams show these attributes in the Curve/Surface elements. The  
> example files also differ, SEDML3.xml has them in 'Plot' but SEDML12  
> and 21 have them in 'Curve'.
>  Which is correct?
>  Cheers
>    Richard
> 
> 
> 
> -- 
> Dr Richard Adams
> Software Development Team Leader,
> Centre For Systems Biology Edinburgh
> University of Edinburgh
> Tel: 0131 650 8285
> email : ric...@ed...
> 
> -- 
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
> 
> 
> 
> ------------------------------------------------------------------------------
> The Palm PDK Hot Apps Program offers developers who use the
> Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> of $1 Million in cash or HP Products. Visit us here for more details:
> http://p.sf.net/sfu/dev2dev-palm
> _______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss

[SED-ML-discuss] log axes on plots or lines

[Richard A. <ra...@st...>]

Hello SEDMLers

  There seems to be some discrepancies as to where the logX/logY/logZ  
attributes should go, from looking in the spec - E.g., section 4.6.2  
text says these attributes go in the Plot2d/3d elements, but the UML  
diagrams show these attributes in the Curve/Surface elements. The  
example files also differ, SEDML3.xml has them in 'Plot' but SEDML12  
and 21 have them in 'Curve'.
  Which is correct?
  Cheers
    Richard



-- 
Dr Richard Adams
Software Development Team Leader,
Centre For Systems Biology Edinburgh
University of Edinburgh
Tel: 0131 650 8285
email : ric...@ed...

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

[SED-ML-discuss] SVN repository structure updated

[Dagmar W. <dag...@un...>]

Hej there,

just to inform everyone that I cleaned and restructured the SVN 
repository on Sourceforge.

So in case anybody is linking to it, you might want to check the links, 
especially for the UML SED-OM and the schema file location.

Best,
Dagmar

Re: [SED-ML-discuss] Change class

[Nicolas Le n. <le...@eb...>]

On 24/06/10 13:36, Dagmar Waltemath wrote:

>>     We could rename to 'xxxElement'? as this doesn't have so much XML
>> connotation.
>>
>
> Hmm, but xxxElement excludes xxxAttributes (from the XML point of view).
> For me, it would imply somehow that only (XML) elements are addressable.
> Could we just call them "Add", "Remove" and "Change" maybe (not
> absolutely convinced myself)?

SED-ML is meant to be used with XML. We address the variables with XPath...

I do not see the problem with the XML connotation. The day SED-ML will work 
with something else (which would first require the existence of a non-XML 
standard for encoding models), the changes in SED-ML will be so drastic 
that the changeThingy will be just a detail.

-- 
Nicolas LE NOVERE, Computational Neurobiology, EMBL-EBI, Wellcome-Trust
Genome Campus, Hinxton CB101SD UK, Mob:+447833147074, Tel:+441223494521
Fax:468,Skype:n.lenovere,AIM:nlenovere,MSN:nle...@ho...(NOT email)
http://www.ebi.ac.uk/~lenov/, http://www.ebi.ac.uk/compneur/, @lenovere

Re: [SED-ML-discuss] mathml schema

[Richard A. <ra...@st...>]

I think we should have a schema defining the MathML that SEDML  
accepts, which is included as part of the spec, for completeness and  
to allow proper validation of documents - if it's just  SBML-MATHml  
plus the  sum/product/min/max constructs, it won't be too arduous to  
modify the SBML-MathML schema.

  Also having  a defined MathML subset for SEDML ( as is written in  
the spec) means that we don't have to stay tied to SBML subset, if we  
don't want to.

Cheers
  Richard

> good question ... and yes if we do schema validation we could just add them
> :) ... or ... if you think it is not worth having in L1V1 we could take it
> out again ...
>
> best
> Frank
>
>> -----Original Message-----
>> From: Richard Adams [mailto:ra...@st...]
>> Sent: Thursday, June 24, 2010 5:34 AM
>> To: sed...@li...
>> Subject: [SED-ML-discuss] mathml schema
>>
>>
>> Hello,
>>   I was wondering if we needed a separate mathml schema document to
>> describe the range of mathml supported by SED-ML?
>>   So, since the sum, min, max, product tags were added have we now
>> diverged from the mathml schema file used by sbml?
>>    Cheers
>>       Richard
>>
>> --
>> Dr Richard Adams
>> Software Development Team Leader,
>> Centre For Systems Biology Edinburgh
>> University of Edinburgh
>> Tel: 0131 650 8285
>> email : ric...@ed...
>>
>> --
>> The University of Edinburgh is a charitable body, registered in
>> Scotland, with registration number SC005336.
>>
>>
>>
>>
> ----------------------------------------------------------------------------
> --
>> ThinkGeek and WIRED's GeekDad team up for the Ultimate
>> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
>> lucky parental unit.  See the prize list and enter to win:
>> http://p.sf.net/sfu/thinkgeek-promo
>> _______________________________________________
>> SED-ML-discuss mailing list
>> SED...@li...
>> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit.  See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
>



-- 
Dr Richard Adams
Software Development Team Leader,
Centre For Systems Biology Edinburgh
University of Edinburgh
Tel: 0131 650 8285
email : ric...@ed...

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

Re: [SED-ML-discuss] mathml schema

[Frank T. B. <fbergman@u.washington.edu>]

good question ... and yes if we do schema validation we could just add them
:) ... or ... if you think it is not worth having in L1V1 we could take it
out again ... 

best
Frank

> -----Original Message-----
> From: Richard Adams [mailto:ra...@st...]
> Sent: Thursday, June 24, 2010 5:34 AM
> To: sed...@li...
> Subject: [SED-ML-discuss] mathml schema
> 
> 
> Hello,
>   I was wondering if we needed a separate mathml schema document to
> describe the range of mathml supported by SED-ML?
>   So, since the sum, min, max, product tags were added have we now
> diverged from the mathml schema file used by sbml?
>    Cheers
>       Richard
> 
> --
> Dr Richard Adams
> Software Development Team Leader,
> Centre For Systems Biology Edinburgh
> University of Edinburgh
> Tel: 0131 650 8285
> email : ric...@ed...
> 
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
> 
> 
> 
>
----------------------------------------------------------------------------
--
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit.  See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss

[SED-ML-discuss] mathml schema

[Richard A. <ra...@st...>]

Hello,
  I was wondering if we needed a separate mathml schema document to  
describe the range of mathml supported by SED-ML?
  So, since the sum, min, max, product tags were added have we now  
diverged from the mathml schema file used by sbml?
   Cheers
      Richard

-- 
Dr Richard Adams
Software Development Team Leader,
Centre For Systems Biology Edinburgh
University of Edinburgh
Tel: 0131 650 8285
email : ric...@ed...

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

Re: [SED-ML-discuss] Change class

[Richard A. <ra...@st...>]

How about xxxNode? xxxEntity? xxxIdentifiableEntity? xxxTarget?

I suppose the key point is that part  of the model is identifiable and  
can therefore be targeted for manipulation.

> Richard Adams wrote:
>> Hi Dagmar,
>>   I thought we'd agreed that the add/remove/change elements were to be
>> subelements of a 'changeXML' parent element.
>
> First, I thought to do that, too. But for RemoveXML a "newXML" attribute
> (as needed in AddXML and ChangeXML) would not be suitable anymore. So I
> thought to have them in parallel instead.
>
>> But this looks fine.
>>    We could rename to 'xxxElement'? as this doesn't have so much XML
>> connotation.
>>
>
> Hmm, but xxxElement excludes xxxAttributes (from the XML point of view).
> For me, it would imply somehow that only (XML) elements are addressable.
> Could we just call them "Add", "Remove" and "Change" maybe (not
> absolutely convinced myself)?
>
>>     Cheers
>>       Richard
>>
>>
>>> Hello all,
>>>
>>> just to summarize the discussion we had on the Change class, and what I
>>> think is the current solution.
>>>
>>> The way that most people would like to go is having different classes
>>> AddXML, RemoveXML and ChangeXML.
>>>
>>> In the class hierarchy that might look like:
>>>
>>> - Change
>>> --- ComputeChange
>>> (to compute a change of a piece of XML addressed by an XPath by whatever
>>> the mathML expression defines)
>>> --- AddXML
>>> (to add a piece of XML as sub-element nested in the element addressed by
>>> an XPath)
>>> --- RemoveXML
>>> (to remove exactly the piece of XML that is addressed by the XPath)
>>> --- ChangeXML
>>> (to replace a piece of XML addressed by an XPath by what is inside the
>>> "newXML" attribute)
>>>
>>>
>>> Could we get rid of the ChangeAttribute class then?
>>>
>>> Does anyone want a different naming of yyXML in case we want to move
>>> away from XML in the future or is there no necessity for that?
>>>
>>> Does anyone want to see examples for how the changes would look like in
>>> a SED-ML XML file?
>>>
>>> If no complaints I will update the XML/UML accordingly.
>>>
>>> Dagmar
>>>
>>> ------------------------------------------------------------------------------
>>> ThinkGeek and WIRED's GeekDad team up for the Ultimate
>>> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
>>> lucky parental unit.  See the prize list and enter to win:
>>> http://p.sf.net/sfu/thinkgeek-promo
>>> _______________________________________________
>>> SED-ML-discuss mailing list
>>> SED...@li...
>>> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>>>
>>>
>>>
>>
>>
>>
>>
>
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit.  See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> SED-ML-discuss mailing list
> SED...@li...
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>
>



-- 
Dr Richard Adams
Software Development Team Leader,
Centre For Systems Biology Edinburgh
University of Edinburgh
Tel: 0131 650 8285
email : ric...@ed...

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

Re: [SED-ML-discuss] Change class

[Dagmar W. <dag...@un...>]

Richard Adams wrote:
> Hi Dagmar,
>   I thought we'd agreed that the add/remove/change elements were to be  
> subelements of a 'changeXML' parent element. 

First, I thought to do that, too. But for RemoveXML a "newXML" attribute 
(as needed in AddXML and ChangeXML) would not be suitable anymore. So I 
thought to have them in parallel instead.

> But this looks fine.
>    We could rename to 'xxxElement'? as this doesn't have so much XML  
> connotation.
>   

Hmm, but xxxElement excludes xxxAttributes (from the XML point of view). 
For me, it would imply somehow that only (XML) elements are addressable.
Could we just call them "Add", "Remove" and "Change" maybe (not 
absolutely convinced myself)?

>     Cheers
>       Richard
>
>   
>> Hello all,
>>
>> just to summarize the discussion we had on the Change class, and what I
>> think is the current solution.
>>
>> The way that most people would like to go is having different classes
>> AddXML, RemoveXML and ChangeXML.
>>
>> In the class hierarchy that might look like:
>>
>> - Change
>> --- ComputeChange
>> (to compute a change of a piece of XML addressed by an XPath by whatever
>> the mathML expression defines)
>> --- AddXML
>> (to add a piece of XML as sub-element nested in the element addressed by
>> an XPath)
>> --- RemoveXML
>> (to remove exactly the piece of XML that is addressed by the XPath)
>> --- ChangeXML
>> (to replace a piece of XML addressed by an XPath by what is inside the
>> "newXML" attribute)
>>
>>
>> Could we get rid of the ChangeAttribute class then?
>>
>> Does anyone want a different naming of yyXML in case we want to move
>> away from XML in the future or is there no necessity for that?
>>
>> Does anyone want to see examples for how the changes would look like in
>> a SED-ML XML file?
>>
>> If no complaints I will update the XML/UML accordingly.
>>
>> Dagmar
>>
>> ------------------------------------------------------------------------------
>> ThinkGeek and WIRED's GeekDad team up for the Ultimate
>> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
>> lucky parental unit.  See the prize list and enter to win:
>> http://p.sf.net/sfu/thinkgeek-promo
>> _______________________________________________
>> SED-ML-discuss mailing list
>> SED...@li...
>> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss
>>
>>
>>     
>
>
>
>   

[SED-ML-discuss] Change class

[Dagmar W. <dag...@un...>]

Hello all,

just to summarize the discussion we had on the Change class, and what I 
think is the current solution.

The way that most people would like to go is having different classes 
AddXML, RemoveXML and ChangeXML.

In the class hierarchy that might look like:

- Change
--- ComputeChange
(to compute a change of a piece of XML addressed by an XPath by whatever 
the mathML expression defines)
--- AddXML
(to add a piece of XML as sub-element nested in the element addressed by 
an XPath)
--- RemoveXML
(to remove exactly the piece of XML that is addressed by the XPath)
--- ChangeXML
(to replace a piece of XML addressed by an XPath by what is inside the 
"newXML" attribute)


Could we get rid of the ChangeAttribute class then?

Does anyone want a different naming of yyXML in case we want to move 
away from XML in the future or is there no necessity for that?

Does anyone want to see examples for how the changes would look like in 
a SED-ML XML file?

If no complaints I will update the XML/UML accordingly.

Dagmar