add-rxn.prop: Hydrogen Sulfide Addition

[wtscott31]

This is an enhancement to the model to properly account for the pathways responsible for the formation of hydrogen sulfide as well as other volatile sulfur compounds during fermentation (see source:https://academic.oup.com/femsyr/article/17/6/fox058/4056150)

Main improvements in this PR:

Try to be as clear as possible: Is it fixing/adding something in the model? Is it an additional test/function/dataset? PLEASE DELETE THIS LINE.

I hereby confirm that I have:

• Tested my code with all requirements for running the model
• Selected develop as a target branch (top left drop-down menu)
• If needed, asked first in the Gitter chat room about this PR

[edkerk]

Reviewing this PR, I realized that we did not have an intuitive and straightforward way to make such curations. To fix this, not just for here, but also for future usage, I wrote a function code/curateMetsRxnsGenes() that can take standardized TSV files as input, and adds the required metabolites, genes and/or reactions, with various warnings and error messages if something is going wrong. See the updated code/modelCuration/addSULnewRxn.m to see how this is run, and notice that the TSV files have some minor changes. This curateMetsRxnsGenes() will likely be incorporated in RAVEN in the near future, but for now is distributed with yeast-GEM.

Then, down to the actual model changes:

• There were three transport reactions (cytoplasm <=> extracellular) that were called "exchange", this was corrected.
• Three real exchange reactions were added, at the moment only allowing excretion.
• If you make a new branch for curating the model, make sure you branch it from the devel branch, not the main branch!

To do (@wtscott31) :

• The gene YOL164W is included, but should probably also be assigned to a particular reaction. If it is not used in any reaction, there is no need to add it to the model.

Note: the model was exported by using the branch from PR #301 and RAVEN from branch PR #396. Due to various generic model changes (identifiers suffices, sorting of entities etc.), it looks there were massive changes in the model files. Commit 031c9ce is the best representative of real model changes related to sulfur metabolism.

[wtscott31]

Gene YOL164W has been determined to regulate bacterially derived sulfatase, which is essential for converting sulfate esters to sulfates (see: https://www.yeastgenome.org/locus/S000005524). However, stoichiometry gets tricky when dealing with any known sulfate esters. Perhaps, the reactions are compound reactions with more than one step. Anyway, since this problem is hard to pin down, we can remove gene YOL164W for now.

…

On Mon, Mar 14, 2022 at 12:17 AM Eduard Kerkhoven ***@***.***> wrote: Reviewing this PR, I realized that we did not have an intuitive and straightforward way to make such curations. To fix this, not just for here, but also for future usage, I wrote a function code/curateMetsRxnsGenes() that can take standardized TSV files as input, and adds the required metabolites, genes and/or reactions, with various warnings and error messages if something is going wrong. See the updated code/modelCuration/addSULnewRxn.m <https://github.com/SysBioChalmers/yeast-GEM/pull/300/files#diff-6c5cc77126e2912009441440379adda23c896c3cc00e8b7bee89629c6cdf3cb7> to see how this is run, and notice that the TSV files have some minor changes. This curateMetsRxnsGenes() will likely be incorporated in RAVEN in the near future, but for now is distributed with yeast-GEM. Then, down to the actual model changes: - There were three transport reactions (cytoplasm <=> extracellular) that were called "exchange", this was corrected. - Three *real* exchange reactions were added, at the moment only allowing excretion. To do ***@***.*** <https://github.com/wtscott31>) : - The gene YOL164W is included <https://github.com/SysBioChalmers/yeast-GEM/pull/300/files#diff-b8d4806f0f3e36f033dcfcca8262d8a83c26463f9ed55687f8b818d6b6e2c50d>, but should probably also be assigned <https://github.com/SysBioChalmers/yeast-GEM/pull/300/files#diff-eed9dffcd56f67bf61d505e72e39071bd4add558e7850083d080c9c1e8c78d51> to a particular reaction. If it is not used in any reaction, there is no need to add it to the model. Note: the model was exported by using the branch from PR #301 <#301> and RAVEN from branch PR #396 <SysBioChalmers/RAVEN#396>. Due to various generic model changes (identifiers suffices, sorting of entities etc.), it looks there were massive changes in the model files. Commit 031c9ce <031c9ce> is the best representative of *real* model changes related to sulfur metabolism. — Reply to this email directly, view it on GitHub <#300 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AISXXYQZEVFNFO4WWTZ5AGLU7ZZQ3ANCNFSM5QJAQS7Q> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[edkerk]

Ok, because also in the article that you cite it shows BDS1 as involved in sulfate ester -> sulfate (Fig 1). But are none of the reactions that you propose to add an example of such a reaction?

[wtscott31]

The only example metabolite mentioned in the article for that GPR is sodium dodecyl sulfate, but I cannot find any KEGG or MetaNetX reactions associated with sodium dodecyl sulfate. This would be helpful to avoid stoichiometry problems as it is a large molecule.

…

On Mon, Mar 14, 2022 at 3:34 PM Eduard Kerkhoven ***@***.***> wrote: Ok, because also in the article that you cite it shows BDS1 as involved in sulfate ester -> sulfate (Fig 1). But are none of the reactions that you propose to add an example of such a reaction? — Reply to this email directly, view it on GitHub <#300 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AISXXYQY4AOZZ65NH7LVYFDU75E5NANCNFSM5QJAQS7Q> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[edkerk]

With commit bbcc58f:

• Merged refactor: reduce dependencies and remove metabolite ID suffixes #301, for simpler model I/O.

With commit ef8eaf6:

• Added a few additional metabolite identifiers (MetaNetX, BiGG, BioCyc).
• Curated existing hydrogen sulfide metabolites for consistent identifiers.
• Renamed 2-oxo-3-sulfanylpropanoate to more recognizable 3-mercaptopyruvate.

Remaining issues (@wtscott31):

• The 3-mercaptopyruvate sulfurtransferase should be corrected. I noticed that dihydrolipoate disulfide (DHLA) was annotated with the KEGG ID for thioredoxin disulfide. Inspecting the cited reference, it seems to refer to Mikami et al. 2011, which suggests that both DHLA and thioredoxin are required in mice. Thioredoxin is currently not part of the 3-mercaptopyruvate sulfurtransferase reaction. Should DHLA (and dihydrolipoate disulfide) be replaced with oxidized & reduced thioredoxin?
• If DHLA should be kept in the reaction above, it does not seems to be recycled: it should also be produced in another reaction, otherwise this pathway is blocked.
• Related to that, there does not seem to be a purpose for the dihydrolipoate-diphosphate reaction: it produces dihydrolipoyl-AMP that is not consumed elsewhere. Considering that this is a reaction without gene association, it should be left out.

[wtscott31]

With commit bbcc58f:

• Merged refactor: reduce dependencies and remove metabolite ID suffixes #301, for simpler model I/O.

With commit ef8eaf6:

• Added a few additional metabolite identifiers (MetaNetX, BiGG, BioCyc).
• Curated existing hydrogen sulfide metabolites for consistent identifiers.
• Renamed 2-oxo-3-sulfanylpropanoate to more recognizable 3-mercaptopyruvate.

Remaining issues (@wtscott31):

• The 3-mercaptopyruvate sulfurtransferase should be corrected. I noticed that dihydrolipoate disulfide (DHLA) was annotated with the KEGG ID for thioredoxin disulfide. Inspecting the cited reference, it seems to refer to Mikami et al. 2011, which suggests that both DHLA and thioredoxin are required in mice. Thioredoxin is currently not part of the 3-mercaptopyruvate sulfurtransferase reaction. Should DHLA (and dihydrolipoate disulfide) be replaced with oxidized & reduced thioredoxin?

If you look at the reference Huang et al. 2017, Fig. 1 ((https://academic.oup.com/femsyr/article/17/6/fox058/4056150) TUM1 gene associated reaction is a required part of sulphur metabolism. It is also cited here:(https://pathway.yeastgenome.org/YEAST/NEW-IMAGE?object=TUM1). I think we should include (1b) [3-mercaptopyruvate sulfurtransferase]-S-sulfanyl-L-cysteine + reduced thioredoxin = hydrogen sulfide + [3-mercaptopyruvate sulfurtransferase]-L-cysteine + oxidized thioredoxin [RN:R12690] for 3-mercaptopyruvate sulfurtransferase instead of the current one.

• If DHLA should be kept in the reaction above, it does not seems to be recycled: it should also be produced in another reaction, otherwise this pathway is blocked.

See answer above.

• Related to that, there does not seem to be a purpose for the dihydrolipoate-diphosphate reaction: it produces dihydrolipoyl-AMP that is not consumed elsewhere. Considering that this is a reaction without gene association, it should be left out.

Yes, it should be left out. See answer above.

[edkerk]

I think we should include (1b) [3-mercaptopyruvate sulfurtransferase]-S-sulfanyl-L-cysteine + reduced thioredoxin = hydrogen sulfide + [3-mercaptopyruvate sulfurtransferase]-L-cysteine + oxidized thioredoxin [RN:R12690] for 3-mercaptopyruvate sulfurtransferase instead of the current one.

That would only be a half-reaction, where the [3-mercatopyruvate sulfurtransferase]- metabolite is the actual gene associated to this reaction. Instead I've now included the whole reaction: TRX1 + 3-mercaptopyruvate --> TRX1 disulphide + pyruvate + hydrogen sulfide.

[wtscott31]

I think that reaction should be acceptable then.