search

SBRG / bigg_models

The BiGG Models website server
http://bigg.ucsd.edu
Other
80 stars 18 forks source link

SBML3 for iMM1415 is too big to load #155

Closed zakandrewking closed 9 years ago

zakandrewking commented 9 years ago

Error: 

bigg2/static/models/iMM1415.xml: 
Malformed XML file: Excessive depth in document: 256 use XML_PARSE_HUGE option, line 92888, column 515

XML_PARSE_HUGE does not fix this

@aebrahim @draeger

draeger commented 9 years ago

Hi Zak,

I don't have access to this file, could you please provide it? On http://bigg.ucsd.edu/models/iMM1415 I see only the polished file, not the original one that seems to cause the trouble. Please correct me if I am wrong.

Thanks Andreas

aebrahim commented 9 years ago

That linked polished file is the one that causes the trouble. On Aug 18, 2015 8:32 PM, "Andreas Dräger" notifications@github.com wrote:

Hi Zak,

I don't have access to this file, could you please provide it? On http://bigg.ucsd.edu/models/iMM1415 I see only the polished file, not the original one that seems to cause the trouble. Please correct me if I am wrong.

Thanks Andreas

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-132434095.

draeger commented 9 years ago

Thanks. I checked this file and it is indeed not valid SBML. In order to prevent such cases, I need all input files for ModelPolisher (can you please provide a ZIP archive for me) and I will then validate each individual output file.

aebrahim commented 9 years ago

The raw file for iMM1415 probably looks something like this: http://drosophi.la/iMM1415_raw.xml (I generated it from JSON but it should be the same)

zakandrewking commented 9 years ago

It's here too:

http://bigg.ucsd.edu/static/models/raw/iMM1415.xml

On Wed, Aug 19, 2015 at 11:34 AM, Ali Ebrahim notifications@github.com wrote:

The raw file for iMM1415 probably looks something like this: http://drosophi.la/iMM1415_raw.xml (I generated it from JSON but it should be the same)

Reply to this email directly or view it on GitHub: https://github.com/SBRG/BIGG2/issues/155#issuecomment-132734882

draeger commented 9 years ago

Got it, thanks! It would be great to have all raw files and to be able to locally run the polishing for testing purposes. Could it be possible to provide one ZIP file with all models from http://bigg.ucsd.edu/static/models/raw/?

draeger commented 9 years ago

The new version 0.9 of ModelPolisher (see https://github.com/SBRG/BioNetView/tree/master/dist/ModelPolisher/0) creates a totally valid SBML file.

nel3 commented 9 years ago

Have we tested all of the sbml files to ensure that they can also be correctly loaded by the cobra toolbox, cobrapy, and other tools such as the tools from other groups?

Nathan E. Lewis

Assistant Professor Department of Pediatrics University of California, San Diego Tel: (858) 997 - 5844 http://lewislab.ucsd.edu/

On Tue, Sep 8, 2015 at 1:03 AM, Andreas Dräger notifications@github.com wrote:

The new version 0.9 of ModelPolisher (see https://github.com/SBRG/BioNetView/tree/master/dist/ModelPolisher/0) creates a totally valid SBML file.

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-138468321.

pillmill commented 9 years ago

All current models in the database loaded and validated without errors using Cobra Toolbox 2.0.5 and SBML Toolbox-4.1.0.

nel3 commented 9 years ago

Thanks!

Nathan E. Lewis

Assistant Professor Department of Pediatrics University of California, San Diego Tel: (858) 997 - 5844 http://lewislab.ucsd.edu/

On Thu, Sep 10, 2015 at 12:29 PM, pillmill notifications@github.com wrote:

All current models in the database loaded and validated without errors using Cobra Toolbox 2.0.5 and SBML Toolbox-4.1.0.

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-139354142.

aebrahim commented 9 years ago

Just curious... how are you loading in the SBML+fbcv2 files from BIGG into the COBRA toolbox? I wasn't aware that there was support for this format in the COBRA toolbox yet.

Or are you using the mat format? (Or am I just very confused)?

On Thu, Sep 10, 2015 at 12:32 PM, nel3 notifications@github.com wrote:

Thanks!

Nathan E. Lewis

Assistant Professor Department of Pediatrics University of California, San Diego Tel: (858) 997 - 5844 http://lewislab.ucsd.edu/

On Thu, Sep 10, 2015 at 12:29 PM, pillmill notifications@github.com wrote:

All current models in the database loaded and validated without errors using Cobra Toolbox 2.0.5 and SBML Toolbox-4.1.0.

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-139354142.

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-139354889.

nel3 commented 9 years ago

also, if they load, are they loading all of the content (including directionality, flux bounds, reaction ids and names, kegg ids, etc. etc.)

Nathan E. Lewis

Assistant Professor Department of Pediatrics University of California, San Diego Tel: (858) 997 - 5844 http://lewislab.ucsd.edu/

On Thu, Sep 10, 2015 at 12:33 PM, Ali Ebrahim notifications@github.com wrote:

Just curious... how are you loading in the SBML+fbcv2 files from BIGG into the COBRA toolbox? I wasn't aware that there was support for this format in the COBRA toolbox yet.

Or are you using the mat format? (Or am I just very confused)?

On Thu, Sep 10, 2015 at 12:32 PM, nel3 notifications@github.com wrote:

Thanks!

Nathan E. Lewis

Assistant Professor Department of Pediatrics University of California, San Diego Tel: (858) 997 - 5844 http://lewislab.ucsd.edu/

On Thu, Sep 10, 2015 at 12:29 PM, pillmill notifications@github.com wrote:

All current models in the database loaded and validated without errors using Cobra Toolbox 2.0.5 and SBML Toolbox-4.1.0.

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-139354142.

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-139354889.

— Reply to this email directly or view it on GitHub https://github.com/SBRG/BIGG2/issues/155#issuecomment-139355180.

draeger commented 9 years ago

And are these files generated with ModelPoliser v0.9?

pillmill commented 9 years ago

Tests were done using the xml files in the latest dump. The toolbox function 'TranslateSBML' was used with the validate option 'on' and the warnings option 'off'. Repeating with warnings 'on', the output was: 'LibSBML has detected what appears to be an SBML Level 3 package ... The information relating to 'fbc' will be saved but cannot be interpreted.'