Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

cobra.io.write_sbml_model does not write reaction.subsystem to xml file #473

Closed
sulheim opened this issue Mar 29, 2017 · 5 comments
Closed

cobra.io.write_sbml_model does not write reaction.subsystem to xml file #473

sulheim opened this issue Mar 29, 2017 · 5 comments
Labels
SBML Related to reading and writing SBML models.

Comments

@sulheim
Copy link

sulheim commented Mar 29, 2017 •
edited
Loading

OS: Windows 10
cobra version: 0.5.11
python version: 3.5.2

Ass the title says, all reactions have a subsystem attribute, however this attribute is not stored when using write_sbml_model(). It is the same with the notes attribute. If set the flag use_fbc_package = False, the subsystem is stored as a note

`import cobra

model = cobra.Model("test model")
reaction = cobra.Reaction("test_reaction")
reaction.subsystem = "TCA cycle"
model.add_reaction(reaction)
cobra.io.write_sbml_model(model, "subsystem_test.xml")
`
subsystem issue.zip
@cdiener
Copy link
Member

cdiener commented Mar 29, 2017

Can confirm. Currently we only write annotations and ignore notes and the somewhat special "subsystem". Since there is no DB in identifiers.org that takes a pathway name directly it would have to go into notes. For instance the SBML 3 AGORA models use the following

      <notes>
        <body xmlns="http://www.w3.org/1999/xhtml">
          <p>SUBSYSTEM: Lysine metabolism</p>
          <p>Confidence Level: </p>
        </body>
      </notes>

@sulheim
Copy link
Author

sulheim commented Mar 30, 2017

Ok. Sounds good.

@JuBra
Copy link
Contributor

JuBra commented Apr 10, 2017

Just as a note, we had a discussion about saving subsystems some time ago in: #229

@cdiener
Copy link
Member

cdiener commented Apr 10, 2017 •
edited
Loading

@JuBra thanks for the cross-link. That is indeed a much more qualified discussion about he topic. Seems complicated. I agree with @aebrahim that the best solution would be using the annotations, however that would require mapping the legacy subsystem to some database. I will look around if I find a suitable ID system for that. The problem with the notes field is that it is not parsed by the SBML3 parser since it is supposed to be consumed by humans as is (thus the HTML formatting). So if I remember previous discussions with Ali correctly cobrapy is not supposed to parse the notes field and the fact that this is done in the SBML 2 parser is more of a hack to enable something annotation-like in the annotation-less SBML 2.

Edit: Upon review there is no good way to map the subsystem to existing MIRIAM resources. Most are simply to incomplete and there are also subsystems such as "sink reactions" which you will never find in those databases. I think the major problem is that SBML has no format for human-specified information that should be machine-readable (it only support machine-machine and machine/human - human)...

@cdiener cdiener added the SBML Related to reading and writing SBML models. label Jul 10, 2017
@cdiener
Copy link
Member

cdiener commented Jul 10, 2017

After some discussion it seems that the SBML maintainers recommend using groups for that feature, so this issue will be superseded by #543.

@cdiener cdiener closed this as completed Jul 10, 2017
@akaviaLab akaviaLab mentioned this issue Feb 26, 2020
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
SBML Related to reading and writing SBML models.
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@sulheim @cdiener @JuBra