phantomas1234
commented
7 years ago @cdanielmachado Given the adoption of the mentioned standards, I am not convinced this is doable (at least in the short-term) and maybe out of scope of what memote will provide? I could see though that the usage of memote-based repositories could be very helpful for curators at biomodels.net if they contain code that reproduces figures from publications (also cobra models can and should be deposited there btw)?
In general, I think we were aiming for a "implementation first, standards later" approach with memote (hope you agree that this makes) in case of missing standards.
When you submit a kinetic model to Biomodels.net, you are supposed to indicate one or more figures from the paper that should be reproducible from the model. The manual curator assigned to your submission runs your model and reproduces the figure from the paper. See for example:
https://www.ebi.ac.uk/biomodels-main/simulation-result.do?uri=publ-model.do&mid=BIOMD0000000565
I think it would be a nice to have something similar done with memote (preferably in an automated way) so that paper reviewers and readers can easily reproduce simulation results from the paper where the model is published.
This would require that the exact experimental conditions (exchange reaction rates, gene deletions, etc), as well as the simulation results, are provided in some standard format together with the model.
There are some standards to describe kinetic model experiments, see for instance:
I am not sure if some of these could be applied to constraint-based models as well. If so, they might be useful, and might be worth further consideration. If not, then let's not create a bunch of new standards. To be honest I find the co-existence of so many standards with apparently similar goals somewhat confusing.