Open ChristianLieven opened 6 years ago
Would we check for only use this database or would we also include others like BiGG?
This is in my opinion no quality criteria for a model. If a model is deposited somewhere it should be annotated in the model information. Basically if you want to apply the model in any commercial context later on you can't share the model under CC0 license (which is required by biomodels and most other repositories). Or is there any repository where you keep the full copyright and can distribute under a non-commercial license?
One case, however, that could be checked is if the model is accessible at all. Here is an example for a published model in form of a password-protected Excel file: http://jb.asm.org/content/190/8/2790.long
@matthiaskoenig I don't see the harm to implement this check as a reminder to contribute to open science. Not as a scored test, but rather as yet another statistic/ warning to make users aware. In the end, it is still up to the model builder to act on it.
Further, I may be misunderstanding the following, as I don't quite see how this stands in the way of a commercial application:
You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission.
In no way are the patent or trademark rights of any person affected by CC0, nor are the rights that other persons may have in the work or in how the work is used, such as publicity or privacy rights.
I do agree, however, that it would be nice to have a reference to a model repository stored within the model annotation. This would further simplify our job of checking a model's availability. I don't think that this type of annotation exists, but then again I'm still not sufficiently familiar with SBML. In any case, I have only seen links pointing to the source in a notes field containing the model description so far.
@draeger Right now we can only check SBML files, but yeah those are the cases in principle that I'd like to make users aware of. Like I wonder, did the editors/ reviewers of that model get an unlocked version and then they published the locked one?
@siddC From a user's, builder's and software dev perspective, I'd appreciate a one-stop-shop for models like releasing python packages to pypi, R packages to CRAN and js packages to npm. Not sure how much work it is currently for a model maintainer to keep all online representations of a model up to the latest release. But as it stands currently there are several, non-overlapping archives: Bigg, minho, biomodels, I'm sure there may be more. Thus, I think a general availability check is a good idea.
@ChristianLieven Frankly, I doubt that any reviewer ever tryed to open the model at all, neither did the editor.
Problem description
Many models are not available/ hidden away on institute website or they live only in the supplements of publications. This check could be a nice reminder to link up all models to a central resource.