Daniela,
the short answer: you cannot
the longer answer is that there is no absolute mutation rate setting. For example, the mutation=CONSTANT assumes that the mutation rate
is the same for all loci, but this is just an assumption because we estimate the parameters always as a scaled version, so Theta is the mutation-scaled population size (some other researcher will call that the population-size-scaled mutation rate), which is (4Ne*mu) or the migration parameter M in migrate is the ’mutation-scaled immigration rate’ (which is m/mu. The other options, like DATA, take the variability in the loci and use that as a weight to scale the parameters. If you do not have many loci with a wide range of variabilities, then this option does not work well. I have not experimented with the OWN option lately, but this also only sets weights for each locus, all these measures set the average mutation rate to 1.0.
If you have a single locus, then none of these options really matter.
Peter