geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
223 stars 40 forks source link

NTR - metabolism: [GDP-mannose biosynthetic process from glucose] #27683

Open rozaru opened 7 months ago

rozaru commented 7 months ago

Please provide as much information as you can:

image

Thank you

Rossana

@sjm41 @ukemi @deustp01


rozaru commented 7 months ago

Missed the following parent term(s): mannose catabolic process GO:0019309

pgaudet commented 7 months ago

@rozaru

Just to make sure I understand:

Are you suggesting that the new term 'GDP-mannose biosynthetic process' is_a mannose catabolic process?

ie you'd consider mannose to mannose-6-phosphate a mannose catabolic process?

rozaru commented 7 months ago

@pgaudet,

I should have mention that this comment is related to another ticket #27682 where I suggested to move "GDP-mannose biosynthetic process from mannose" under "mannose catabolic process". If the new term "GDP-mannose biosynthetic process from glucose" is created it would go at the same level as "GDP-mannose biosynthetic process from mannose" is in the ontology.

This is the current ontology for mannose metabolic process:

mannose metabolic process |GDP-mannose biosynthetic process from mannose |mannose biosynthetic process |mannose catabolic process | |glycolytic process from mannose through fructose-6-phosphate | |__mannose to fructose-6-phosphate catabolic process

This is my proposition:

mannose metabolic process |mannose biosynthetic process |mannose catabolic process | |glycolytic process from mannose through fructose-6-phosphate | |mannose to fructose-6-phosphate catabolic process | |GDP-mannose biosynthetic process from mannose | |GDP-mannose biosynthetic process from glucose (new term)

It is a bit difficult to decide what constitutes a catabolic or a biosynthetic process when the output and input belong to multiple pathways. For "GDP-mannose biosynthetic process from mannose": From the point of view of GDP-mannose this term cover a biosynthetic process From the point of view of mannose this term cover its conversion into GDP-mannose which could been seen as a catabolic process.

rozaru commented 7 months ago

@pgaudet I realised that the commend above: Missed the following parent term(s): mannose catabolic process GO:0019309 is a mistake, as the potential new term is not part of the mannose metabolic process. Sorry about this.

pgaudet commented 7 months ago

+[Term] +id: GO:0141199 +name: GDP-mannose biosynthetic process from glucose +namespace: biological_process +def: "The chemical reactions and pathways resulting in the formation of GDP-mannose from glucose." [PMID:21273394, PMID:9287314] +xref: MetaCyc:PWY-5659 +intersection_of: GO:0009058 ! biosynthetic process +intersection_of: has_primary_input CHEBI:4167 ! D-glucopyranose +intersection_of: has_primary_output CHEBI:21168 ! GDP-mannose +relationship: starts_with GO:0004347 ! glucose-6-phosphate isomerase activity +property_value: term_tracker_item "https://github.com/geneontology/go-ontology/issues/27683" xsd:anyURI +created_by: pg +creation_date: 2024-04-26T08:43:27Z

pgaudet commented 7 months ago

Note that I have removed the activities from the LD of the siblings 'GDP-mannose biosynthetic process from fructose-6-phosphate' and 'GDP-mannose biosynthetic process from mannose', since these were not strictly necessary to make the terms uniquely defined; only inputs are outputs are sufficient.

I moved the MFs to assertions.

@ukemi @cmungall please let me know if this is not right.

Thanks, Pascale

rozaru commented 7 months ago

@pgaudet I just notice a small detail, but shouldn't +intersection_of: has_primary_output CHEBI:21168 ! GDP-mannose be +intersection_of: has_primary_output CHEBI:57527 GDP-α-D-mannose(2−) ?

pgaudet commented 7 months ago

Thanks @rozaru !

I fixed this, and for the parent & siblings.

sjm41 commented 2 months ago

@rozaru @ukemi @deustp01 @pgaudet

Following the NTR in this ticket, we end up with this:

GDP-mannose biosynthetic process (GO:0009298) [Reactome:R-HSA-446205]
    |__GDP-mannose biosynthetic process from fructose-6-phosphate (GO:0061729)
    |__GDP-mannose biosynthetic process from glucose (GO:0141199)  [MetaCyc:PWY-5659]
    |__GDP-mannose biosynthetic process from mannose (GO:0061728)

Looking again now, I think the NTR actioned here (GO:0141199) should be merged with the pre-existing term 'GO:0061729 GDP-mannose biosynthetic process from fructose-6-phosphate.

Here's my reasoning:

  1. If you look at Rossana's nice image in the original comment, the 'from fructose-6-phosphate' term is just describing the last 3 steps of the new 5-step 'from glucose' term.
  2. Fig 1 and the accompanying text of PMID:9287314 Rossana mentions (shown below) says that "Man-6-P can be formed in two ways; either directly from mannose using hexokinase, or derived from the pathway of Glc → Glc-6-P ↔ Fru-6-P ↔ Man-6-P". That is, there is one route from mannose, or another route from glucose via F6P.
  3. Fig 1 of PMID:16339137 (used as the source for the 'from fructose-6-phosphate' (GO:0061729) term) also shows that the fructose-6-phosphate originates from Glucose, and again shows the same two routes - also shown below.
  4. There are only 2 EXP annotations to the 'from fructose-6-phosphate' term (to human PMM2 and MPI), and they were both added recently by Rossana!

That said, there are still a few issues:

Maybe we can talk about this in the weeds call tomorrow afternoon?

Screenshot 2024-09-18 at 13 03 21

Screenshot 2024-09-18 at 13 03 47

Screenshot 2024-09-18 at 13 05 26

Screenshot 2024-09-18 at 13 06 15

sjm41 commented 1 week ago

From pathways2go discussion today:

Decision: Merge the new term (GDP-mannose biosynthetic process from glucose GO:0141199) into the pre-existing term (GDP-mannose biosynthetic process from fructose-6-phosphate GO:0061729), keeping the name and def of the pre-existing term. (That is, revert to what we had before, but add the MetaCyc:PWY-5659 xref to GO:0061729). @pgaudet Can you take care of that please?

Justification: Although papers describe the second GDP-Man synthesis as starting from glucose, the first two steps (catalysed by HK and PGI) are best seen as being part of glycolysis. The resulting F(6)P is at a branch point - it can either continue to be metabolised in glycolysis (via PFK) or can be diverted into GFP-Man synthesis (using PMI, aka MPI). Though not perfect, this is the best (simplest) way of modularising and representing the information within GO-BP terms and GO-CAMs. (And it also mirrors the way Reactome models the situation.)

Other consequences: The way we are thinking about this specific example will hopefully help us with subsequent NTRs, and for reviewing other existing metabolic BP GO terms that appear to overlap or are nested in some way.... (e.g. based on the above arguments, there's a case for obsoleting "GO:0061619 glycolytic process from mannose through fructose-6-phosphate" and just keeping the "GO:0061611 mannose to fructose-6-phosphate catabolic process" term in order to connect mannose to core glycolysis.) We will assess and review additional cases as we encounter them and aim to be consistent and establish guidelines!