Now the way that mcmcgan operates with demographic parameters is more flexible than before. The program allows specifying which parameters are inferrable and which are not. This should simplify the implementation of more demographic models with other parameters (growth rates, migration rates, times, etc...).
Also, the user should have the flexibility to specify the values for the parameters to infer (right now the parameters are predefined and given in the params_dict variable, in genobuilder.py)