[cellml-discussion] curation status of models in the repository

[James Lawson]

Andrew Miller wrote:
> James Lawson wrote:
>> Hi David,
>>   
> ...
>> Because of the way this model is built, getting the units to match would
>> require drastically remaking it, and I don't think I'd know how to do
>> it. I have some ideas, but only because this is a particularly simple
>> model, with 3 differential equations and two algebraic expressions. In a
>> larger model I would have absolutely no idea where to start.
>>
>> So what is the solution to this? Only code up models in CellML that are
>> perfect? Because technically, this model isn't valid CellML
>>   
> Nothing in the CellML specification says that valid CellML models cannot 
> have mathematically-inconsistent units in the equations, so it is still 
> valid CellML.

Okay sure, I'm just going from what the validator said (i.e. that the
units mismatch means that the model isn't valid CellML 1.0)

> 
> One of the aims of the repository is for it to represent models as they 
> were published in journals (or to at least represent the model that the 
> author used to derive their conclusions, if this happens to differ from 
> the model actually described).
> 
> If an existing paper has issues like:
> * Units are inconsistent.
> * Mass, charge, water balance etc... doesn't hold.
> * Model lacks long-term stability of some kind
> and so on, then we should still put the model in the repository, but we 
> should annotate the model in some way to warn potential users of the 
> model about these issues (and therefore allow someone creating a 
> derivative model to improve on the problems).
> 
> Of course, it is often the case that papers don't fully state the units 
> of everything, in which case we need to insert our own units, and any 
> problems with the units that aren't inherent in the paper should be 
> fixed before the model is uploaded.
> 
> Best regards,
> Andrew
> 
>> Kind regards,
>> James
>>
>>
>>
>> David Nickerson wrote:
>>   
>>> Hi all,
>>>
>>> I am just after a bit of clarification on the use of the stars with the 
>>> models in the model repository. As I understand the model curation (from 
>>> http://www.cellml.org/repository-info/info), one star means the model is 
>>> at curation level 1, two stars for level 2, and three for level 3 
>>> curation. For the model simulation tools, there is one star if the model 
>>> loads in and runs, 2 if the model gives the right results, and 3 stars 
>>> if the results have been rigorously verified.
>>>
>>> The first question I have is if the stars next to the "Standard" link 
>>> represent the model curation level while those next to specific tools 
>>> correspond to the running of simulations of that model with that tool? 
>>> In other words, it is not very obvious what all the stars are implying 
>>> when looking at a model's page in the repository.
>>>
>>> Looking at the models with two stars (thanks to Tommy's new filter tool 
>>> on the main page), I see they generally have two stars for Standard, 
>>> PCEnv and COR but no stars for JSim. If the "Standard" stars do reflect 
>>> the curation level of the model, then my second question is how are the 
>>> units are being checked (as required for level 2 curation) without using 
>>> JSim? I was under the impression that JSim was the only tool capable of 
>>> throughly checking all the units for consistency in a model, but perhaps 
>>> there is something else? And if so, should it not also be displayed 
>>> along with PCEnv, COR, and JSim?
>>>
>>>
>>> Thanks,
>>> David.
>>>
>>>     
>> _______________________________________________
>> cellml-discussion mailing list
>> cellml-discussion at cellml.org
>> http://www.cellml.org/mailman/listinfo/cellml-discussion
>>   
> 
> _______________________________________________
> cellml-discussion mailing list
> cellml-discussion at cellml.org
> http://www.cellml.org/mailman/listinfo/cellml-discussion