As suggested by a reviewer, it'd provide helpful context to say what sort of data was used to fit the demographic models, so users can guess how what aspects of the data simulations are expected to recapitulate. As discussed in the meeting, it'd be nice to add this to the descriptions of the models. I don't think this should be extensive, and we do a little of this already; along the lines of "inferred from site frequency spectrum data using dadi" or "inferred by simulation-based deep learning using genotype matrices"; for the real details one will have to look at the original paper.
In fact, it might be too hard to actually provide a sufficiently useful and accurate summary; so maybe this is not useful? But if we think it is, then we need to
update existing models
add guidance to the "how to add a model" documentation
As suggested by a reviewer, it'd provide helpful context to say what sort of data was used to fit the demographic models, so users can guess how what aspects of the data simulations are expected to recapitulate. As discussed in the meeting, it'd be nice to add this to the descriptions of the models. I don't think this should be extensive, and we do a little of this already; along the lines of "inferred from site frequency spectrum data using dadi" or "inferred by simulation-based deep learning using genotype matrices"; for the real details one will have to look at the original paper.
In fact, it might be too hard to actually provide a sufficiently useful and accurate summary; so maybe this is not useful? But if we think it is, then we need to