Dear Vanina,
I looked at the run/data and it seems that my version 4.4.4 has no problem with it, before I dig out version 4.4.1 could you check whether
the locus 7 data leads to the problem [because node 7 was running locus 7], this locus has no variability (all repeats are 6), either remove the locus or (for test purposes change two individuals to a heterzogyte and run again). Does it work? Please let me know.
Of course, migrate should work with an invariant microsat locus, but it seems that on my mac 4.4.4 works despite the problems with the parmfile,
I suggest to remove the line:
print-tree=BEST
this will (a) not be insightful with msat data, will need a lot of time, and is ill formatted
I also suggest to revisit the priors for MIG because the upper bounds together with Theta suggest an 4Nm>50*250, using a prior for MIG in the same range of Theta (for msats) seems still excessive but more in line.
Peter