search

geneontology / pathways2GO

Code for converting between BioPAX pathways and Gene Ontology Causal Activity Models (GO-CAM)
8 stars 0 forks source link

Make appropriate pathways cyclical #285

Closed suzialeksander closed 6 months ago

suzialeksander commented 1 year ago

On the managers call, @thomaspd mentioned the pathways that should be cycles are not because of limitations in the BioPax (steps). It would be nice if these could be fixed on import.

@dustine32 does this sound feasible? What (if anything) would you need from curators- list of affected pathways? Not sure if this can be done on import/code side or should be part of the planned manual curation.

dustine32 commented 1 year ago

@suzialeksander Yes, before actually trying this, it does sound feasible as long as we emit only one instance of each location-specific molecule (e.g., ChEBI) class. The code currently matches up only outputs of each pathway step with the inputs of their next steps. If we expand this to match inputs to inputs and outputs to outputs, this should solve branched and cyclical pathways.

Branched pathway example: PWY3O-6635 superpathway of phosphatidate biosynthesis Cyclical pathway example: PWY3O-94 superpathway of TCA cycle and glyoxylate cycle

suzialeksander commented 1 year ago

That sounds like what Paul was saying. I think that would be ok, as long as it didn't get confused if some same molecules are in/outputs of different steps- so on the TCA cycle, "H2O + acetyl-CoA" as input with the output "H+ coenzyme A" occurs both from malate synthase and citrate synthase.

suzialeksander commented 11 months ago

Update: @dustine32 will try to work on this after the Reactome2GO meeting this week

suzialeksander commented 6 months ago

This is a task SGD curators are working on.