search

geneontology / pathways2GO

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

occurs_in for Yeast Pathways is incorrect #273

Closed suzialeksander closed 9 months ago

suzialeksander commented 10 months ago

Newly found issue: gomodel:YeastPathways_PWY-2201 / Pathway: folate transformations I

SHM1 and MIS1 (Mitochondrial C1-tetrahydrofolate synthase) should both be mitochondrial, but there are no "occurs_in mitochondria", they're all cytosol. I'm looking at the BioPax 3, I don't see where the CC comes from except 

     <bp:cellularLocation>
      <bp:CellularLocationVocabulary rdf:ID="CellularLocationVocabulary2368">
        <bp:term rdf:datatype="http://www.w3.org/2001/XMLSchema#string">cytosol</bp:term>
        <bp:xref>
          <bp:UnificationXref rdf:ID="UnificationXref2369">
            <bp:id rdf:datatype="http://www.w3.org/2001/XMLSchema#string">GO:0005829</bp:id>

@dustine32 Is there a way to check if this is an annotation issue? Are there any models that have more than one CC, for example? Does the code apply the cellularLocation to everything in the model or just specified entities- I'm not incredibly fluent in BioPax.

dustine32 commented 10 months ago

@suzialeksander As far as I can tell, this is an annotation issue. Yes, the BioPAX files have cytosol for all proteins but 3 models have plasma membrane (GO:0005886) for certain small molecules: PWY3O-94 PYRIMID-RNTSYN-PWY TCA-EUK-PWY

For the example proteins SHM1 and MIS1, I can't really tell on the pathway.yeastgenome.org pages (aside from having "mitochondrial" in the labels) what the cellular locations are. Like, is there a field that stores location in YeastPathways that then is meant to be exported out to the BioPAX?

suzialeksander commented 9 months ago

SGD will curate these manually. Closing.