SweeD selection sweep regions per populations

[Anik Du]

Hi Pavlos,

Thank you very much for developing SweeD. I am running SweeD on individual chromosome with 99 samples that come from 9 populations. So, each population has 11 individuals. I have run the SweeD like this and got the output with a user-defined 99% percentile significance threshold. 

From the population structure and phylogenetic analysis, I see that there are 7 population subgroups according to geographic origin. Do you think, I should run SweeD based on individual population when there is such strong population structure? If I run SweeD on each population (11 individuals), will the analysis lack power to detect sweeps due to low sample size?

If I run SweeD on all populations combined, then how can I separate the populations those have the selective sweeps and those do not have?

Thank you.