Closed mephenor closed 5 years ago
Basically now we need to create the escher maps right? In order to do so we can either insert manually all the reactions in the subsystem (lipopolysaccharide biosynthesis, for now) and connect all the metabolites (since many are shared among the reactions), which would be quite time consuming, or write a program that creates a JSON file with all these connections already done, is this right?
Escher maps is the next step, though I won't have time to do this today, as I'm seriously lagging behind on work for another project. As described in the Escher paper, there should be a Python API, so I'm not sure if it's necessary to generate the JSON files, might be possible to have it somehow interact directly with the information from the model, though I haven't had the time to have a look at how this is supposed to be done
I just had a look into the Escher Documentation, unfortunately it doesn't seem to be easily possible. I couldn't find something like an "addReaction" method, the only way they describe is to open the map in a browser or notebook and then adding reactions by hand. We can ask Mr. Dräger on Thursday if there's a way to do this programmatically or if we have to do it ourselves.
Using the model with added group information splitting the model into slices with regard to subsystems should be doable by using the group-member idref to extract relevant reactions and metabolites.
Using COBRApy on the generated slices we should be able to get the JSON representation for only this subsystems. Alternatively it might be possible to generate the pathway maps directly in a notebook using Escher's Python API.
I will have a look at this on monday.