TABLE 2.

Accuracy of GEBV, prediction bias, and rate of inbreeding (mean of 100 replicates) for different genotyping scenarios for fish in C.

	h2 = 0.1				h2 = 0.3
rg	Random	T1_1B1_1	T1_2B1_2	Top1_1	Random	T1_1B1_1	T1_2B1_2	Top1_1
Accuracy of GEBV
0.2	0.362	0.515	0.534	0.245	0.512	0.635	0.656	0.371
0.5	0.389	0.522	0.544	0.284	0.525	0.641	0.661	0.395
0.8	0.436	0.548	0.564	0.363	0.555	0.65	0.666	0.474
SE	0.005	0.004	0.004	0.005	0.004	0.003	0.002	0.005
Prediction bias
0.2	0.961	0.59	0.506	1.232	0.976	0.528	0.465	1.423
0.5	1.045	0.624	0.544	1.324	1.023	0.554	0.484	1.428
0.8	1.207	0.763	0.656	1.434	1.142	0.645	0.573	1.527
SE	0.013	0.005	0.005	0.024	0.008	0.003	0.003	0.017
Rate of inbreeding
0.2	0.729	0.872	0.851	0.643	0.858	0.881	0.901	0.762
0.5	0.842	0.838	0.887	0.646	0.944	0.913	0.963	0.81
0.8	0.761	0.93	0.887	0.674	0.931	0.952	0.975	0.734
SE	0.022	0.030	0.028	0.017	0.031	0.028	0.037	0.027

Genotyping of fish in C: random (Random); selection of the phenotypically best and worst fish from each sample of 20 fish (T1_1B1_1); selection of fish from one of the two top fish and one of the two bottom fish, where the top and bottom fish were the best and worst fish, respectively, from each sample of 20 fish (T1_2B1_2); and selection of the phenotypically best fish from each sample of 20 fish (Top1_1). r_g is the genetic correlation between the trait measured in B and C. h² is the heritability of the trait simulated. The scenarios used selective genotyping of top fish for selection candidates in B, and 100% of selected females were kept in 2 consecutive years. Standard errors (SE) are the average values for different r_g. GEBV, genomic breeding values; C, commercial environment; B, breeding environment.