Invalid annotations?

[smoretti]

Hi

some annotations don't look to be valid in the Sco-GEM.xml file:

• ec-code: according to identifiers.org ec-code annotations must contain a single EC number. For cases with multiple EC numbers, I think you have to duplicate the annotations. Don't know really how to deal with OR And not sure what you mean with the comma: "2.7.1.11, 2.7.1.56, 2.7.1.144". AND or OR ? e.g. "https://identifiers.org/ec-code/"1.2.4.2 and 1.8.1.4 and 2.3.1.61"" "https://identifiers.org/ec-code/1.1.1.1 or 1.1.99.1" "https://identifiers.org/ec-code/2.7.1.11, 2.7.1.56, 2.7.1.144"

• kegg.reaction: according to identifiers.org kegg.reaction must contain a single KEGG reaction id. I think you have to duplicate the annotations. Again not sure what you mean with the comma. e.g. "https://identifiers.org/kegg.reaction/['R02075', 'R08836']"

• doi/PMID: is not valid according to identifiers.org registry for doi Better to use https://registry.identifiers.org/registry/pubmed for PMID e.g. "https://identifiers.org/doi/PMID: 16554727" should be https://identifiers.org/pubmed/16554727

[sulheim]

Hey! Thanks for pointing these errors out. The gene annotation would be the place were the and / or relation is defined, so the ec-numbers should be duplicated annotations (i.e. one annotation per enzyme). The reaction 45DOPA should only be annotated to R08836. I'll add these to #105

[smoretti]

Hi it looks Sco-GEM 1.3.0 (xml version) still contains uncommon annotations:

• Some ec-code annotations are not valid according to the pattern defined in identifiers.org ec-code annotations must contain a single EC number. e.g. "https://identifiers.org/ec-code/"1.2.4.2 and 1.8.1.4 and 2.3.1.61"" "https://identifiers.org/ec-code/3.4.11.1 or 3.4.11.2" should be

              "https://identifiers.org/ec-code/1.2.4.2"
              "https://identifiers.org/ec-code/1.8.1.4"
              "https://identifiers.org/ec-code/2.3.1.61"
              "https://identifiers.org/ec-code/3.4.11.-"

• Some kegg.reaction annotations are not valid according to the pattern defined in identifiers.org kegg.reaction must contain a single KEGG reaction id. e.g. "https://identifiers.org/kegg.reaction/['R09768', 'R05645']" should be (if the comma separator means a logic AND)

            "https://identifiers.org/kegg.reaction/R09768"
            "https://identifiers.org/kegg.reaction/R05645"

Also it looks the parameters/fluxes "minus_inf" and "plus_inf" are defined but never user. The same for the chemical species "M_g1p_B".

[edkerk]

Now resolved in #145, let us now if you spot any other issues! The minus_inf and plus_inf bounds are added by cobrapy to any SBML model by default.

[smoretti]

Thanks Everything looks fine my side

Best