Mutation rate per generation or per year?

[jguhlin]

Working with a wild pop using tskit trees (tsinfer + tsdate) and my units are in years. Does Phlash expect it to match or should I just use per generation?

[terhorst]

Sorry, I missed this! phlash does not directly use tskit trees (yet...) other than to extract the raw sequence data from them. Similar to PSMC, the units that you specify for the mutation rate will govern the time scaling of the resulting estimates. In particular if you know the per generation mutation rate, then the time axis will be in "generations ago". Hope this helps.