True path violations under NON proteinogenic amino acid metabolic process

[ValWood]

arginine biosynthetic process via ornithine (GO:0042450) arginine catabolic process to ornithine (GO:0019547) L-methionine biosynthetic process from L-homoserine via cystathionine (GO:0019279) L-methionine biosynthetic process from homoserine via O-acetyl-L-homoserine and cystathionine (GO:0019280) L-methionine biosynthetic process from homoserine via O-succinyl-L-homoserine and cystathionine (GO:0019281)

[pgaudet]

You mean they should be under 'proteinogenic'? If that's the case, you reported this 3 weeks ago here: https://github.com/geneontology/go-ontology/issues/28367

Should we close one as redundant?

[ValWood]

doh, in my head i was thinking this was different. I did know some were not under proteinogenic that should be, but I didn 't realise they were actually in the wong place.

[deustp01]

Are we tripping here over the fact that some proteinogenic amino acid biosynthetic pathways include intermediates that are also amino acids, like ornithine, but that are not themselves proteinogenic, so depending on the molecule you look at the process is proteinogenic (arginine) or not (ornithine), which is why (I still think) imposing this distinction at the level of amino acid biosynthesis is dangerous. Impose it instead at the level of the processes that consume arginine or ornithine or glutamate ...

[pgaudet]

No, the problem is the ph/.3 forms, see the other ticket: https://github.com/geneontology/go-ontology/issues/28367#issuecomment-2203300882

[ValWood]

The ph forms is an issue, but I think @deustp01 is correct here. These have a weird parentage and it was on my list to report it.

For example L-methionine biosynthetic process from L-homoserine via cystathionine (GO:0019279)

has

Some children of "methionine biosynthetic process" are descendants of the associated intermediate metabolic process (which has a none proteinogenic amino acid parent ) This seems to be the case for all of these examples above.

[ValWood]

Isn't the problem still going to exist even after the zwitterion issue is resolved, because homoserine is an input ?

Is this the correct way to describe a pathway? I thought MFs have inputs and pathways have starts and ends?