cytosolic proton in non-cytosolic reactions

[mihai-sysbio]

Description of the issue:

Reaction MAR05259 is pumping out a proton in cytosol, according to its reaction equation. Because of this, all of the reaction associated genes end up also being associated to the cytosol.

Expected feature/value/output:

MAR05259 would be expected to have all metabolites in one compartment (lyososme) by using MAM02039l.

Current feature/value/output:

MAR05259 produces MAM02039c.

I hereby confirm that I have:

• [ ] Tested my code on my own computer for running the model
• [x] Done this analysis in the master branch of the repository
• [x] Checked that a similar issue does not exist already

[haowang-bioinfo]

It does appear that the proton should be lysosomal version instead of cytosol one (MAM02039c), which probably was added via charge balancing?

[JonathanRob]

Not sure whether it was added during charge balancing or some other reason, but I agree that the compartment should be changed as @mihai-sysbio suggests.

[haowang-bioinfo]

Yes, go ahead with the change if it's mass- and charge-balanced.

[mihai-sysbio]

It is hard to find the change so way back before the yml-based workflow was in place. As far as I can see in the yml file, the change was introduced as part of the commit fe8d50e46d3056324eb20c0e03dfc0bbae42b8d1 which was the last commit before merging 1.2 via #148. It does look like the change was part of the rebalancing process since it shows up in a file named rebalance_modelChanges_rxns.tsv.

[mihai-sysbio]

I would also want to mention that there is a related reaction MAR05258 taking place fully in the cytosol.

[haowang-bioinfo]

@mihai-sysbio the reaction MAR05259 most likely is not the only case of mixing cytosol proton with rxns in another compartment. Could a thorough check be conducted for this?

[mihai-sysbio]

@mihai-sysbio the reaction MAR05259 most likely is not the only case of mixing cytosol proton with rxns in another compartment. Could a thorough check be conducted for this?

The cytosol proton is involved in over 1k reactions, but I did look at all reactions where the coefficient was large, ie. not +-1, and could not find any similar cases.

[haowang-bioinfo]

Apart from MAR05259, there are two additional similar cases MAR02041 (from ER) and MAR04758 (Mitochondrion) that have: i) cytosol proton involved; ii) but all other metabolites in another compartment.

[haowang-bioinfo]

The cytosol proton in these three reactions (MAR05259: lysosome; MAR02041: ER; MAR04758: mitochondrion) may be fixed together.

[mihai-sysbio]

While I see how MAR02041 should be fixed by using a different proton, I'm not sure what to say about MAR04758, especially after looking at the reactions facilitated by gene FECH. I am more tempted to simply delete MAR04758 since it would otherwise duplicate MAR01044.

[haowang-bioinfo]

I am more tempted to simply delete MAR04758 since it would otherwise duplicate MAR01044.

good idea - MAR04758 and MAR01044 are indeed duplicated reactions

[mihai-sysbio]

Actually, it looks like MAR02041 should also be carefully merged with MAR01459. @Hao-Chalmers I would need some help with this.

[haowang-bioinfo]

will look into this

[haowang-bioinfo]

After looking into the MAR04758 and MAR02041 reactions, it is found that each is duplicated with another as detailed below, as well as planned actions:

• MAR04758 <=> MAR01044

rxn ID	MAR04758	MAR01044
compartment	mitochondrion	mitochondrion
cytosolic H+	yes	no
GPR	ENSG00000066926	ENSG00000066926
PMID	11175906; 11215517; 11853551; 11947230; 1624416; 182145; 2310748; 3196293; 3702737; 406931; 6425295; 7309736; 8818224; 9712849; 9808757	7983009
EC	4.99.1.1	4.99.1.1
source	HMR	Recon
planned action	remove	keep

• MAR02041 <=> MAR01459

rxn ID	MAR02041	MAR01459
compartment	ER	ER
cytosolic H+	yes	no
GPR	ENSG00000025423 or ENSG00000086696 or ENSG00000108786 or ENSG00000132196 or ENSG00000149084 or ENSG00000198189 or ENSG00000204228	ENSG00000132196
PMID	19027824	12732193
EC	1.1.1.62; 1.1.1.64	1.1.1.62
source	HMR	Recon
planned action	fix cytosolic H+ and keep	remove

@mihai-sysbio @JonathanRob what do you think?

[mihai-sysbio]

Sounds good to me, I'll go ahead and implement this.

[mihai-sysbio]

I think this issue can be closed, since #283 was merged.