Table 2.

Comparison of haplotype frequencies using distance$d$, maximal absolute difference between frequencies $\text{increment}$, and first rank with a relative deviation larger than 0.05, $\rho$

Data	Remark	$d$	$∆$	$\rho$
First artificial population	Integer-valued genotype numbers	$1.3\times {10}^{-4}$	$9.04\times {10}^{-7}$	None
	Integer-valued genotype numbers and NMDP codes	$0.11\pm 0.02$	$\left(4\pm 1\right)\times {10}^{-3}$	$14\pm 6$
Second artificial population	Hapl-o-Mat – Population	$0.200$	$3.1\times {10}^{-3}$	$1$
	Arlequin – Population	$0.202$	$3.2\times {10}^{-3}$	$1$
	Arlequin – Hapl-o-Mat	$0.027$	$2.2\times {10}^{-4}$	$106$
Real genotype data	Hapl-o-Mat – Arlequin	$0.072\pm 0.002$	$\left(9\pm 2\right)\times {10}^{-4}$	$41\pm 23$

The observables were computed on basis of original and estimated haplotype frequencies. For the first artificial population, where we compared Hapl-o-Mat to population data, the column “Remark” indicates details of construction. For the other two genotype data sets, it indicates the sets of haplotype frequencies that are compared to each other, e.g. “Hapl-o-Mat – population” means haplotype frequencies obtained from Hapl-o-Mat were compared to original population haplotype frequencies