Repository navigation
MAM01026 and MAM00348 have the same name, an issue for VMH #375
Description
hello,
Could you please have a look at the '2 metabolites found' here:
https://www.vmh.life/#allsearch/oretn
MAM01026 4-oxo-13-cis-retinoate is 'oretn'
MAM00348 13-cis-oxo-retinoate is '13_cis_oretn'
the name is similar/identical but only MAM00348 corresponds to the -cis stereo form at carbon 13 (I recently created the corresponding chebi for that chemical, should hopefully be validated soon as CHEBI:190136)
MAM01026 is the compound that can be mapped to CHEBI:134186 all-trans-4-oxoretinoate, both have the same mol file actually
As a consequence, I send this message to VMH via their contact form, and I hope you agree
Hello Team,
I am currently doing chemical curation work for MetaNetX.org.
It sounds like the compounds with abbreviation '13_cis_oretn' and 'oretn' have the same name (13-Cis-Oxo-Retinoic Acid, and 4-Oxo-13-Cis-Retinoate), while only '13_cis_oretn' corresponds to the structure with a -cis stereo at carbone 13. 'oretn' should maybe be redefined as 'all-trans-4-oxoretinoate' such as CHEBI:134186, right?
Thank you in advance and best regards,
Anne
Anne Niknejad
Senior biocurator
SIB | Swiss Institute of Bioinformatics Unil, Department of Ecology and Evolution - Quartier Sorge, bâtiment Biophore - 1015 Lausanne t: +41 21 692 42 21 - f: +41 21 692 41 65 Anne.Niknejad@sib.swiss - https://www.sib.swiss
I hereby confirm that I have:
- Tested my code on my own computer for running the model
- Done this analysis in the
mainbranch of the repository - Checked that a similar issue does not exist already
Metadata
Metadata
Assignees
Labels
No labels
Activity
JonathanRob commentedon Mar 28, 2022
Thanks @ANiknejad for identifying this. It seems that both metabolites (
MAM01026andMAM00348) are supposed to correspond to the same 4-oxo-13-cis form, though some of the reactions in which they are involved in Human-GEM may need to be investigated (e.g.MAR08703). The all-trans-4-oxo form is actually already present in the model as theMAM01031metabolite.Therefore, to address this issue, I suspect that we will need to do the following:
MAM01026andMAM00348metabolites (remove one from the model).MAM01031should be used instead.edkerk commentedon Jul 11, 2026
Thanks for the careful report. Digging in, this turned out to be the reverse of what the naming suggests, so I did an adversarial check against the databases and the reactions before changing anything.
Both metabolites are the same compound, and it is 4-hydroxy, not 4-oxo. MAM01026 was conflated: it carried two MetaNetX IDs pointing at different stereoisomers (MNXM37697 = 13-cis, MNXM6902 = all-trans, CHEBI:134186), which is exactly why its structure looked all-trans while the model used it as cis. But the deeper issue is the oxidation state:
So the formula and reactions are consistent with 4-hydroxy; only the name and cross-refs said 4-oxo. Changing the formula to true 4-oxo would have unbalanced all four reactions, so the fix is to correct the name.
What was done:
All modified reactions remain mass and charge balanced. One thing to note: the true 4-oxo-13-cis-retinoate (the major isotretinoin metabolite) is now not represented in the model; adding it along with the 4-hydroxy -> 4-oxo step, mirroring the all-trans pathway (MAR06702/06703), would be a sensible follow-up.
fix: MAM01026 is 4-oxo-13-cis-retinoate; merge duplicate and correct …
edkerk commentedon Sep 28, 2026
Follow-up to #1044, in #1105 (open): The retinoate 4-hydroxylation and 4-oxo GPRs now follow single-isoform human data (CYP2C9, CYP2B6, CYP2C19, CYP3A5 and CYP26A1 added where shown; CYP1B1 and CYP4A11 left out), and CYP26C1 (new gene, UniProt Q6V0L0) is added to the all-trans and 9-cis 4-hydroxylations and 4-oxo steps (RHEA:51984, 75851, 75847, 75855; PMID:14532297) and the 13-cis 4-hydroxylation (PMID:29476041). Rhea ids are added, the lumped MetaNetX ids are removed from the split hydroxylations, and 4-hydroxy-9-cis-retinoate (MAM20091) gets CHEBI:139253, MNXM1107686 and LMPR01090053. Details and evidence are in the comment on #1044.
edkerk commentedon Oct 5, 2026
Fixed in Human 2.1.0.