A comparison of statistical methods for genomic selection in a mice population

BMC Genomic Data

Table 2 Summary statistics ^* pertaining to phenotypic data ^ employed in cross validation**

Trait	Split	N	Training size			Phenotypes training		Phenotypes testing
Trait	Split	N	Min	Ave	Max	Mean	SD	Mean	SD
W6W	within	1925	1059	1061.9	1066	−0.155	1.96	−0.188	1.95
WGS	within	1917	1056	1059.6	1068	0.001	0.04	0.002	0.04
BL	within	1840	1013	1017.9	1035	−0.002	0.40	−0.006	0.40
%CD8+	within	1407	774	778.6	785	0.010	4.34	0.122	4.36
CD4+/CD8+	within	1403	772	774.5	781	0.003	0.07	0.004	0.07
W6W	across	1925	1059	1067.6	1081	−0.162	1.95	−0.180	1.96
WGS	across	1917	1057	1063.3	1076	0.001	0.04	0.002	0.04
BL	across	1840	1014	1022.3	1030	−0.002	0.40	−0.005	0.40
%CD8+	across	1407	775	780.9	791	−0.149	4.23	0.319	4.43
CD4+/CD8+	across	1403	774	782.2	799	0.003	0.07	0.003	0.07

*N = total number of phenotypic records. Minimum, average and maximum size of training set (Min, Ave and Max) and mean and standard deviation (SD) of the adjusted records considered in training and testing sets (averaged across replicates).^**Traits considered: weight at 6 weeks (W6W), weight growth slope (WGS), body length (BL), percentage of CD8+ cells (%CD8+), ratio between CD4+ and CD8+ cells (CD4+/CD8+). Split = splitting strategy in cross-validation (within or across-family).

ISSN: 2730-6844