No biopax "control" class in biomodels and getting pathway information

[Rashid Saadman]

Hello,

I'm currently a PhD candidate at the University of Cincinnati and I have been trying to use rbiopaxparser(R package) to parse biopax models that I got from biomodels (https://www.ebi.ac.uk/biomodels-main/). I wanted the ACTIVATION/INHIBITION data(adjacency matrix) from the models for pathways and as far as I have read in the documenation for biopax, there has to be a "control" class which I don't get from biopax models in the biomodels database. How do I get the ACTIVATION/INHIBITION data?

Also, how can I get the actual pathway names from the biopax models in biomodels database? They don't have a NAME identifier for the pathway and it's encoded based on MIRIAM.

Thank you


Rashid Saadman Karim

[Igo R]

There is nothing wrong with BioPAX in this regard - 'control' (BioPAX Level2 class) and 'Control' (and Catalysis, Modulation subclasses in BioPAX Level3) are still there, in the specification.

This does not require every data provider generate all the available in the BioPAX ontology classes and properties every time, alright.

You're dealing with BioPAX instance data (models of pathways/interactions), such as from BioModels, BioPAX Level2 I suppose, where there might be no 'control' objects indeed (I haven't checked), which is possibly ok, as long as the authors meant that.

They exported to BioPAX what they wanted or could. There might be other semantic problems, such as with identifiers, controlled vocabularies, etc., - critical or not so (you can tell some is you'd run it through the BioPAX Validator, or summarize with Paxtools app, etc.)