I am working with mtDNA control region sequences, 1 locus, and comparing 4 dolphin populations. I am selecting a subsample of 50 individuals. I'm having problems defining inheritance scaler, I try to modify the value to 1.00, since I'm working with mitochondrial DNA, but the program gives an error, when I click on Y, it doesn't modify the parameter values. Some other parameters cause problems. And when I try to run replicas, the program stops working.
I am trying to follow the methodology of a paper that works with this region in another species of cetacean, the authors used these parameters: 100 independent replicates of the Markov chain scheme: each replicate employed a static heating scheme and involved 15 short chains (dememorization: 10 000 genealogies; recorded genealogies: 100; sampling increment: 100) and 3 long chains (dememorization: 10 000 genealogies; recorded genealogies: 1000; sampling increment: 100).
I was starting with 20 replicas, but the program stops working. Furthermore, the papers I read with cetaceans use maximum likelihood strategy, but reading the MIGRATE manual, I understood that for this type of data I have, Bayesian inference would be better. I would really appreciate help with these analyses, I am new to these analyzes and it is difficult to solve these problems.
Below are the files from two attempt to run the program. First, using default values, and second, trying to modify some parameters.
Thank you very much,
Brenda.