Table 5.

Estimates of Type I Error for Two-Marker Global Haplotype Test

Scenarios	Test	Two Offspring		Four Offspring
		With Parents	Without Parents	With Parents	Without Parents
1	DC Test	0.050	0.051	0.051	0.049
2	NDC Test	0.052	0.053	0.049	0.051
3	DC Test	0.049	0.048	0.050	0.052
4	NDC Test	0.053	0.054	0.051	0.052
5	DC Test	0.052	0.048	0.050	0.049
	NDC Test	0.037	0.034	0.042	0.039
Admixturea,b	DC Test	0.052	0.053	0.051	0.052
6	NDC Test	0.049	0.052	0.051	0.053
	DC Test	0.034	0.032	0.038	0.035
Admixturea,c	NDC Test	0.052	0.053	0.054	0.054

The simulation was based on 5000 replicate of 250 families with MAF set at 0.2 for X-linked QTL and at 0.3 for an autosomal QTL, marker haplotype frequency set as {0.2, 0.3, 0.1, 0.4}, and total variance (V) set at 40, with heritability of X-linked QTL at 0.1. The null hypothesis assumes no association for haplotypes H₀, H₁, and H₂ (D₀ = 0, D₁ = 0, and D₂ = 0).

^aEqual admixture of families drawn from subpopulations, with p = 0.2, marker haplotype frequency distribution {0.25, 0.25, 0.25, 0.25} in one population and p = 0.5, marker haplotype frequency distribution {0.7, 0.1, 0.1, 0.1} in the other subpopulation.

^bAdmixture of subpopulations simulated under scenario 5. We show only the DC test result, because the NDC test is conservative under scenario 5.

^cAdmixture of subpopulations simulated under scenario 6. We show only the NDC test result, because the DC test is conservative under scenario 6.