Is it possible that degree of freedom differs for every STR marker in human data, while the number of population is fixed for all alleles? can this be possible?

Similar questions and discussions