TABLE 1

Estimated percentage of phenotypic variance associated with markers

			Analytical results^a		Simulated results
h²	m²(r)^{^d}	Percent α	Selection bias not included	Selection bias included	Average value over 100 simulations^b	Standard error of the average^{^c}
0.15	0.3 (0.226)	5	0.017	0.059	0.059	0.003
		20	0.029	0.085	0.087	0.004
	0.5 (0.146)	5	0.043	0.088	0.082	0.004
		20	0.060	0.113	0.114	0.004
	0.7 (0.082)	5	0.075	0.119	0.118	0.004
		20	0.094	0.143	0.142	0.004
	0.9 (0.053)	5	0.111	0.150	0.153	0.006
		20	0.128	0.173	0.168	0.005
0.3	0.3 (0.226)	5	0.058	0.103	0.100	0.004
		20	0.077	0.128	0.132	0.005
	0.5 (0.146)	5	0.129	0.166	0.165	0.005
		20	0.144	0.187	0.180	0.005
	0.7 (0.082)	5	0.200	0.228	0.218	0.006
		20	0.207	0.248	0.244	0.006
	0.9 (0.053)	5	0.265	0.288	0.280	0.005
		20	0.269	0.308	0.302	0.005
0.45	0.3 (0.226)	5	0.111	0.150	0.148	0.005
		20	0.128	0.173	0.179	0.005
	0.5 (0.146)	5	0.217	0.243	0.242	0.005
		20	0.223	0.263	0.266	0.005
	0.7 (0.082)	5	0.313	0.334	0.337	0.007
		20	0.315	0.353	0.358	0.005
	0.9 (0.053)	5	0.405	0.424	0.431	0.006
		20	0.405	0.443	0.453	0.006

Comparison of the estimated percentage of phenotypic variance associated with markers obtained ^afrom expectations based on an analytical approach (bias caused by the selection of the markers in the index taken into account or not), and ^b from the average estimated value over 100 simulations.
↵^c The standard error of the average value obtained by simulation was used to determine a confidence interval at the significant level of 5% and to test the adequacy between calculations and simulations. The size of the population is 300, and there are five QTLs of equal effects.
h², heritability of the trait; m², percentage of genetic variance associated with markers; α, type I level risk.
↵^d r indicates the corresponding rate of recombination between marker and QTL.