mutation rate

[ferran]

Dear Joao,

I am a little bit confussed on the units for the mutations rate. For a
set of independent SNPs (humans), should it be set to the human mut
rate 10 -8?

many thanks

ferran

[Joao Sollari Lopes]

Dear Ferran,

following the legend in the prior files, there are 3 options to be used

0 - zero mutation;
1 - lognormal distribution: (mean of mean(log10); stdev of mean(log10); mean of Sdev(log10); stdev of stdev(log10). Stdev truncated at 0.
2 - normal distribution: (mean of mean; stdev of mean; mean of Sdev; stdev of stdev. Stdev truncated at 0.

Let's focus on option 2.

If you want to set a fixed rate of 1e-8 to all locus, use:
2 1e-8 0 0 0

if you are not sure about 1e-8 and want it to vary per simulation (i.e. the mutation rate varies per simulation in the same way as Ne, t and m vary), use [recommended option for 1 locus dataset]:
2 1e-8 1e-9 0 0

if you want the mutation rate to vary between loci (but keep a mean of 1e-8), use:
2 1e-8 0 1e-9 0

if you want the mutation rate to vary between loci and are not sure about its mean being 1e-8 [recommended option for 2 locus dataset], use:
2 1e-8 1e-9 1e-9 0

if you want to add a bit more stochasticity to the mutation rate, you can set all the values to be non-zero:
2 1e-8 1e-9 1e-9 1e-10

Now, regarding the mutation rate units, it is number of mutations per locus per generation (as described in the user guide). If you have a mutation rate in different units you have to convert them to the ones used by the software

Joao