MEMOTE reports my model is rubbish :(

[Stephen Chapman]

Hi all,

I have converted a human model to mouse model by changing the GPR's, and added 5 reactions, 4 metabolites and 2 new genes, and then updated the model to SBML level 3 and FCS . I was happy with the results too, as I could predict an experiment. I decided to check the model using MEMOTE and the report can be found here

Basically the results say my model is not encoded in SBML and is not valid for FCS, (But i'm sure it is as I updated the model using:

cobra.io.sbml.write_sbml_model("model.xml").

What bothers me the most is that the model is being reported as:

      stoichiometric consistency = 0 %

      low metabolite annotation: (I have KEGG metabolite ID's)

      low reaction annotation (i have used reaction EC numbers)

      low gene annotation (I have used ENSEMBLE ID's)

and the overall score is very low at 34%

Can anyone please help me to improve these scores?

Thank you kindly

Stephen

[Amit Kugler]

Hello Stephen and everyone,

I also encountered the same issue. I downloaded the model (xml) from BiGG database and test it on MEMOTE, without any modifications.

All other parameters (Mass Balance, Charge Balance, Metabolite Connectivity) are fine, except Stoichiometric Consistency that shows 0.0%.

Best regards,

Amit