SysBioChalmers / Human-GEM

The generic genome-scale metabolic model of Homo sapiens
https://sysbiochalmers.github.io/Human-GEM-guide/
Creative Commons Attribution 4.0 International
97 stars 40 forks source link

Ammonia/Ammonium Duplicate Reactions #665

Closed Devlin-Moyer closed 1 year ago

Devlin-Moyer commented 1 year ago

The Problem:

There are several reactions that are completely identical except one uses NH4+ (MAM02579) and the other uses NH3 (MAM02578) + H+:

ID Reaction Genes
MAR01515 H2O + methylamine + O2 --> formaldehyde + H2O2 + NH4+ AOC1 or AOC3 or AOC2
MAR08529 H2O + methylamine + O2 --> formaldehyde + H+ + H2O2 + NH3 AOC1 or AOC3 or AOC2
MAR00039 1,3-diaminopropane + H2O + O2 --> 3-amino-propanal + H2O2 + NH4+ AOC1 or AOC3 or AOC2
MAR07992 1,3-diaminopropane + H2O + O2 --> 3-amino-propanal + H+ + H2O2 + NH3 AOC1 or AOC3 or AOC2
MAR00096 3-ureidoisobutyrate + 2 H+ + H2O --> CO2 + D-3-amino-isobutanoate + NH4+ UPB1
MAR04819 3-ureidoisobutyrate + H+ + H2O --> CO2 + D-3-amino-isobutanoate + NH3 UPB1
MAR02027 cysteine + H2O <=> H+ + H2S + NH4+ + pyruvate CTH
MAR04302 cysteine + H2O --> 2 H+ + H2S + NH3 + pyruvate CTH
MAR02006 2 ATP + CO2 + H2O + NH4+ --> 2 ADP + carbamoyl-phosphate + 3 H+ + Pi CPS1
MAR03873 2 ATP + CO2 + H2O + NH3 --> 2 ADP + carbamoyl-phosphate + 2 H+ + Pi CPS1

Uniprot lists all reactions associated with AOC1, AOC2, AOC3, UPB1, CTH, and/or CPS1 with NH4+ rather than NH3.

Proposed Changes:

Since MAR02027 (NH4+) is reversible but MAR04302 (NH3) is irreversible, Uniprot says that reaction should be irreversible, most similar reactions already in Human-GEM involve NH3 (MAM02578) rather than NH4+ (MAM02579), I think it makes sense to remove all of the NH4+ versions and keep all of the NH3 versions. MAR05127: NH4+ [c] ⇔ H+ [c] + NH3 [c] allows ammonium and ammonia to freely interconvert (which is reasonable; that reaction is spontaneous in aqueous solutions), so it's not really contradicting Uniprot to keep the NH3 versions instead of the NH4+ versions, and that would make the proposed changes a little simpler:

feiranl commented 1 year ago

Agree! please go ahead and make these changes!

Devlin-Moyer commented 1 year ago

Fixed in #677