TristanLefebure

#R script
#April 12 2020, TLefebure

#Do we generate an artificial correlation by regressing Pi_s and Pi_N / Pi_S estimated on the same dataset?

#A function to sample polymorphic site on a Ps gradient
#The Pi_N / Pi_S is stable, but Pi_s varies
#Very too naive, also these are not true Pi estimates (nucleotide diversity),
#more like theta_watterson (number of segregatting sites)