Add RECON2 to database

[matthiaskoenig]

Hi all, the bigg model database only contains a very old RECON from around 10 years ago, i.e., Recon1 which nobody is using any more. It would be great if the latest versions of RECON2 could be added to the BiGG model database. This is especially important because the latest human reconstructions are not available as SBML, but only as matlab files.

See here https://vmh.uni.lu/#downloadview

Could you add the following:

11.05.2015: Recon 2.04 - major update of gene-reaction associations (GPRs). Gene identifiers were updated to Build 38. GPRs were refined and extended.

Adding Recon 2 to BiGG would provide extremely valuable to the community, because there would be a a standard representation of the model (SBML) available and the reactions and species would be linked to the universal species and reactions. Also the information provided by model polisher would be very valuable to the community.

In addition, does anybody know where the knowledge base/repository underlying Recon2 is. This was advertised as a "community effort", but I could not find any community knowledge base or github repository where the content of RECON 2 is managed? There has to be more than just a matlab file somewhere?

Best Matthias

[draeger]

Please be a little patient: Soon, a totally new version of Recon coming out and will be in BiGG.

[matthiaskoenig]

I just saw an updated version here, i.e., RECON2.2 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4896983/

@draeger Where is the community database/knowledge base to contribute to Recon2 ? Or is this a "closed community effort"?

[zakandrewking]

It would be great to get new human models into BiGG, but this is a pretty big undertaking given the complicated history of these models and their divergence from RECON1. Therefore, we might have to wait for the release of our upcoming reconstruction platform where these changes can be made collaboratively by the model authors

[matthiaskoenig]

@zakandrewking sounds good. Is there any alpha-/beta- version of the reconstruction platform for testing available (some github repository from which I can build it?) Desperately looking for some usable reconstruction platform for metabolic networks, but all the available things don't do the job (or are only web based without opportunity for self hosting, not open source, or have not sufficient support for SBML). Would be great if I could test this !

[zakandrewking]

Not yet, but I can add you to our beta testing group as soon as we get there

[zakandrewking]

I guess things can change pretty quickly!

http://bigg.ucsd.edu/models/Recon3D

[matthiaskoenig]

Thanks, this is great.

Unfortunately, the ENSG ensemble identifiers as well as all the mappings to uniprot are missing from the SBML. These are both provided in the original publication and crucial to work with the model. For instance the ENSG identifiers are used as identifiers for the gene associations. Could you please add the ENSG annotations to the geneProducts as well as the uniprot annotations.

Also this looks strange: http://bigg.ucsd.edu/models/Recon3D/genes/0

Best M

[matthiaskoenig]

Also how is this model related to the publication?? The BiGG model has

Metabolites | 5835
Reactions | 10600
Genes | 2248

whereas the paper reports:

3,288 open reading frames (representing 17% of functionally annotated human genes), 13,543 metabolic reactions involving 4,140 unique metabolites, and 12,890 protein structures.

[zakandrewking]

Do you mind opening a new issue for these concerns? On Tue, Feb 27, 2018 at 12:40 AM Matthias König notifications@github.com wrote:

Also how is this model related to the publication?? The BiGG model has

Metabolites | 5835 Reactions | 10600 Genes | 2248

whereas the paper reports:

3,288 open reading frames (representing 17% of functionally annotated human genes), 13,543 metabolic reactions involving 4,140 unique metabolites, and 12,890 protein structures.

— You are receiving this because you modified the open/close state.

Reply to this email directly, view it on GitHub https://github.com/SBRG/bigg_models/issues/283#issuecomment-368788358, or mute the thread https://github.com/notifications/unsubscribe-auth/ABMUYE8tU9nQFMD-vUJwjqp6IsoCigF4ks5tY7-bgaJpZM4RkPBQ .

[matthiaskoenig]

Opened separate issues for the single points: https://github.com/SBRG/bigg_models/issues/287 https://github.com/SBRG/bigg_models/issues/288 https://github.com/SBRG/bigg_models/issues/289 Best M