Potential issue with netplotlib not grabbing speciesIds

[kirichoi]

Hi, I just tried with another model (I attach it) and netplotlib didn't put the species names on it.

I took it from here using the select tool.
https://pdmap.uni.lu/minerva/?x=18261&y=5073&zoom=9&searchCoordinates=18130,4689,4895,8&

model_a.xml.zip

Originally posted by @carlosvega in #35 (comment)

[kirichoi]

This is bit odd. I tried to look into the model using Antimony but it does not like it at all.
I am getting these messages:

line 510: (layout-20808 [Error]) The value of the 'layout:reference' attribute of a <generalGlyph> must be of the 'id' of an existing element in the <model>.
Reference: L3V1 Layout V1 Section 3.11
 The <generalGlyph> with the id 'modification_25' has a reference 'modification_residue_0' which is not the id of any element in the model.

I also noticed that the model is using sbml-layout, which is not supported by netplotlib just yet. netplotlib was intended to generate network diagrams from models without layout information. The support is coming through libsbml-draw, however, which should be soon.

[carlosvega]

The support is coming through libsbml-draw, however, which should be soon.

That's great. Thanks for your work. Could it be a way to disregard the layout information and continue drawing the diagram?

The with the id 'modification_25' has a reference 'modification_residue_0' which is not the id of any element in the model.

This might be because the model is exported from a bigger model. Here you can see that only the selected part was exported. But in this case the issue reference is there, but on a multi element.

<multi:speciesFeatureValue id="modification_residue_0" multi:value="minerva_modification_type_GenericProtein_Residue_null_4"/>

[kirichoi]

It might be a good idea to add a function to bypass using layout information. I will take a look.
The error message indeed seems to be coming from missing references.