Scripts to implement a model-based demographic inference for Solanum chilense to assess the support of three demographic models using a wide range prior parameter distributions for effective populations size (Ne), gene-flow and divergence times parameters using Approximate Bayesian Computation approach and NLR data.
Model 1: independent divergence of the north, coastal and mountain groups from the central group.
Model 2: same as model 1 but the highland group diverges from the coastal group.
Model 3: same as model 1 but the coastal group diverges from the highland group.
Divergence processes were set to occur from three different unsampled (i.e. ghost) populations.
The gene-flow among populations from different groups was set as symmetric migration only from populations of the central group with the others (i.e. no migration between north-mountain, north-coast and mountain-coast populations). Therefore, we have seven migration rate parameters: three different migration rates between population from different group (between group migration) and four rates for within-group migration (each group is represented as an island model). Between-group migration reates were constrained to be smaller than within-group gene-flow. Within a group, all populations were assumed to have the same population size for limiting the number of parameters in the models.
