virtualmetabolichuman / agora

The aim of this repository is to maintain, refine, and expand the AGORA collection of genome-scale metabolic reconstructions for human gut microbes through the use of issues. Please note that pull requests will be automatically closed.

AGORA

This repository is dedicated to the genome-scale metabolic reconstruction of human gut microbes (Magnusdottir et al, Nat Biotech, 2017).

The aim of this repository is to allow the user to post issues regarding current content, missing information, and (suggested) expansions.

We will review the issues and address them.

Please ensure to provide literature-evidence for suggested changes or expansions.

Please note that this is a project funded by research grants. Consequently, addressing issues may take time, depending on the complexity of the issue as well as the extent of supporting information provided by the user.

PLEASE NOTE THAT PULL-REQUESTS WILL BE AUTOMATICALLY CLOSED WITHOUT MERGING, as VMH.life already implements version control via a database implementation.

Please cite: Magnusdottir, S., Heinken, A., Kutt, L., Ravcheev, D.A., Bauer, E., Noronha, A., Greenhalgh, K., Jager, C., Baginska, J., Wilmes, P., Fleming, R.M.T., Thiele, I., "Generation of genome-scale metabolic reconstructions for 773 members of the human gut microbiota", Nature Biotechnology, 35(1):81-89 (2017).

https://www.nature.com/articles/nbt.3703

agora's Issues

Flux Variability Analysis min-max mix up in Bifidobacterium_longum_infantis_ATCC_15697

Hi! I have been trying to simulate individual strains, Bifidobacterium longum infantis ATCC 15697 in this case, in BacArena and looking at the flux variability during the simulation with fluxVarSim(). The only media that is in the environment is 10% of essential media and 1mM glc_D. I have attached an image of the results, and the FVA seems to have a mix up between min-max. Does anyone know why this happens? Please let me know if I need to provide any clarification.

EX_for(e) 991.4599 0.0000
EX_fru(e) 0.0000 0.0000
EX_fuc_L(e) 0.0000 0.0000
EX_gal(e) 0.0000 0.0000
EX_gcald(e) 0.0000 0.0000
EX_gchola(e) 0.0000 0.0000
EX_glc_D(e) 0.0000 -610.0665
Thank you for your help!

[AGORA 1.03] Mitochondrion compartment in Clostridium difficile?

Hi there,

While looking at differences between AGORA 1.02 and 1.03, I noticed 4 new compartments in the latter:

Clostridium_difficile_CD196.xml:      <compartment metaid="m" id="m" name="Mitochondrion" constant="false"/>
Clostridium_difficile_NAP07.xml:      <compartment metaid="m" id="m" name="Mitochondrion" constant="false"/>
Clostridium_difficile_NAP08.xml:      <compartment metaid="m" id="m" name="Mitochondrion" constant="false"/>
Clostridium_difficile_R20291.xml:      <compartment metaid="m" id="m" name="Mitochondrion" constant="false"/>

Each of those compartments only contains a single species: hydrogenphosphate (HO4P)

Clostridium_difficile_CD196.xml:      <species metaid="M_pi__91__m__93__" id="M_pi__91__m__93__" name="hydrogenphosphate" compartment="m" hasOnlySubstanceUnits="false" boundaryCondition="false" constant="false" fbc:charge="-2" fbc:chemicalFormula="HO4P">
Clostridium_difficile_NAP07.xml:      <species metaid="M_pi__91__m__93__" id="M_pi__91__m__93__" name="hydrogenphosphate" compartment="m" hasOnlySubstanceUnits="false" boundaryCondition="false" constant="false" fbc:charge="-2" fbc:chemicalFormula="HO4P">
Clostridium_difficile_NAP08.xml:      <species metaid="M_pi__91__m__93__" id="M_pi__91__m__93__" name="hydrogenphosphate" compartment="m" hasOnlySubstanceUnits="false" boundaryCondition="false" constant="false" fbc:charge="-2" fbc:chemicalFormula="HO4P">
Clostridium_difficile_R20291.xml:      <species metaid="M_pi__91__m__93__" id="M_pi__91__m__93__" name="hydrogenphosphate" compartment="m" hasOnlySubstanceUnits="false" boundaryCondition="false" constant="false" fbc:charge="-2" fbc:chemicalFormula="HO4P">

<species metaid="M_pi__91__m__93__" id="M_pi__91__m__93__" name="hydrogenphosphate" compartment="m" hasOnlySubstanceUnits="false" boundaryCondition="false" constant="false" fbc:charge="-2" fbc:chemicalFormula="HO4P">
        <notes>
          <body xmlns="http://www.w3.org/1999/xhtml">
            <p>ChEBIID:43474</p>
          </body>
        </notes>
        <annotation xmlns:sbml="http://www.sbml.org/sbml/level3/version1/core">
          <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dcterms="http://purl.org/dc/terms/" xmlns:vCard="http://www.w3.org/2001/vcard-rdf/3.0#" xmlns:vCard4="http://www.w3.org/2006/vcard/ns#" xmlns:bqbiol="http://biomodels.net/biology-qualifiers/" xmlns:bqmodel="http://biomodels.net/model-qualifiers/">
            <rdf:Description rdf:about="#M_pi__91__m__93__">
              <bqbiol:is>
                <rdf:Bag>
                  <rdf:li rdf:resource="http://identifiers.org/hmdb/HMDB00973"/>
                  <rdf:li rdf:resource="http://identifiers.org/inchi/InChI=1S/H3O4P/c1-5(2,3)4/h(H3,1,2,3,4)/p-2"/>
                  <rdf:li rdf:resource="http://identifiers.org/kegg.compound/C00009"/>
                  <rdf:li rdf:resource="http://identifiers.org/pubchem.compound/1004"/>
                </rdf:Bag>
              </bqbiol:is>
            </rdf:Description>
          </rdf:RDF>
        </annotation>
      </species>

This metabolite is only used in a single reaction (the same for each model):

<reaction metaid="R_IVCOASadp" id="R_IVCOASadp" name="Isovaleryl-CoA synthetase (ADP-forming)" reversible="true" fast="false" fbc:lowerFluxBound="FB1N1000" fbc:upperFluxBound="FB4N1000">
        <notes>
          <body xmlns="http://www.w3.org/1999/xhtml">
            <p>Confidence Level: 0</p>
          </body>
        </notes>
        <listOfReactants>
          <speciesReference species="M_atp__91__c__93__" stoichiometry="1" constant="true"/>
          <speciesReference species="M_coa__91__c__93__" stoichiometry="1" constant="true"/>
          <speciesReference species="M_isoval__91__c__93__" stoichiometry="1" constant="true"/>
        </listOfReactants>
        <listOfProducts>
          <speciesReference species="M_adp__91__c__93__" stoichiometry="1" constant="true"/>
          <speciesReference species="M_ivcoa__91__c__93__" stoichiometry="1" constant="true"/>
          <speciesReference species="M_pi__91__m__93__" stoichiometry="1" constant="true"/>
        </listOfProducts>
        <fbc:geneProductAssociation>
          <fbc:geneProductRef fbc:geneProduct="G_645462__46__3__46__peg__46__2161"/>
        </fbc:geneProductAssociation>
      </reaction>

• I tried to investigate but couldn't get a clue about the reason those compartment, metabolite or reaction are present. What's the story here? I guess it's something that wasn't properly curated out, but I'm curious about why such a compartment would be present in the original model in the first place. It was not present in AGORA 1.02.
• There is no reaction that consume hydrogenphosphate. Doesn't it mean it will accumulate in the system? Couldn't this be a problem in some simulations?

Thank you in advance for your help.

Best regards,
−Nils

[AGORA 1.03] Lower bound for "demand for Biotin" sink reaction is set to 0.1 on some models

Hi and thank you for providing AGORA,

I noticed some strange interactions with biotin (btn) for some of the models provided in AGORA 1.03, which made me investigate. There are 140 models in AGORA 1.03 that contain the following sink reaction for biotin:

<reaction metaid="R_DM_btn" id="R_DM_btn" name="Demand for biotin" reversible="false" fast="false" fbc:lowerFluxBound="…" fbc:upperFluxBound="…">

of which:

• 136 have lower and upper bounds of 0 and 1000
• 4 have lower and upper bounds of 0.1 and 1000

Those 4 models are:

Clostridium_sporogenes_ATCC_15579.xml
Faecalibacterium_prausnitzii_M21_2.xml
Faecalibacterium_prausnitzii_SL3_3.xml
Lactobacillus_fermentum_ATCC_14931.xml

Technically, this lower bound on a sink reaction means that the model cannot grow without having access to at least 0.1 mmol/human/day of biotin (either exchanged from the medium, or produced internally). I initially thought this was because biotin was reported to be an essential metabolite for those 4 bacteria, but when looking at the Supplementary Table provided, there are 32 bacteria for which biotin is noted as essential, and Lactobacillus_fermentum_ATCC_14931 is not even one of them. Apparently such a lower bound was not present in AGORA 1.02.

What's the reason for having a 0.1 mmol/human/day lower bound on this sink reaction in those specific 4 models?

Best regards,
−Nils

Downloading an AGORA model

Hi,

Was wandering whether it's possible to download a metabolic model reconstruction from the AGORA repository.
I aim at using this model for comparing with my models, via MEMOTE. If so, how do I do it?

A following question - are all models in AGORA written in MATLAB format?

Thank you in advance!

Alon

Question about the release date of AGORA 2

Hi,

First of all, thank you for developing and maintaining this valuable database.
I really appreciate efforts of your team.

AGORA 1.03 supports 818 species, so I hope that the number of supported species will increase.
I expect this hope will be fulfilled in AGORA 2, which supports 1,644 species.

I wonder when these AGORA 2 DB will be released.

Best regards,
Young-Ho

Clostridium sp M62/1 Reference Genome

As described in “Clostridium sp M62/1” xml file, this model use txid411489 as reference.
However, I found "Clostridium sp M62/1" taxon id is 411486 in NCBI. NCBI Link

Which one is correct?

Licensing

I would like to deposit an exact copy of the models that I'm using for a study on figshare or similar place in order to get a DOI for them. However, neither here nor on https://www.vmh.life/ nor in the publication do I find any information with regard to the license under which these models are distributed. Should I assume them to be public domain (CC-0)?

Update AGORA genomes download to include most recent version

The AGORA genomes fasta files download "contains the fasta files for each microbe present in the AGORA collection." But it only contains the 773 microbes of v1 (or v1.01?).

I understand that the new genomes in v1.02 can be downloaded manually, but having them available in the bulk download would facilitate access to and analysis of the most recent models.