I have used GTR+G in R using PhyloSim to simulate the evolution of nucleotide sequences along a perfect star topology (I'm doing some work on homoplasy and branch attraction).
G was set to have four categories, and a shape parameter of 0.334437.
At the end of the simulation, when I fit the simulated sequences to the star tree to get branch lengths, I have a more-or-less normal distribution of branch lengths.
Should I expect the distribution of branch lengths to be normal? Or should I expect it to reflect the gamma distribution I set with the shape parameter in the simulations?
I'm just having trouble getting my head around what I should expect to see.
Thank you to anyone who can help.