Validation with single-step SNPBLUP shows that evaluations can continue using a single mean of genotyped individuals, even with multiple breeds

Table 7 Mean validation correlations of 100 replicates for bodyweight at 7 days and 35 days (BW7 and BW35), for each method of selective genotyping (RND or TOP)

Genotyping rate	J-factor method^a	BW7^b		BW35^b
Genotyping rate	J-factor method^a	RND	TOP	RND	TOP
100%^c	Duenk et al. [14]	0.16 (0.058)		0.26 (0.060)
100%^c	ONE^c	0.17 (0.055)		0.26 (0.061)
75%	NONE	0.14 (0.059)	0.13 (0.055)	0.23 (0.059)	0.28 (0.056)
	ONE	0.14 (0.061)	0.14 (0.061)	0.24 (0.061)	0.26 (0.061)
	EXP	0.14 (0.060)	0.14 (0.059)	0.24 (0.061)	0.26 (0.061)
	OBS	0.14 (0.071)	0.14 (0.059)	0.23 (0.055)	0.26 (0.066)
50%	NONE	0.13 (0.063)	0.10 (0.048)	0.21 (0.064)	0.25 (0.057)
	ONE	0.13 (0.057)	0.11 (0.056)	0.21 (0.064)	0.21 (0.065)
	EXP	0.12 (0.057)	0.11 (0.056)	0.21 (0.062)	0.21 (0.064)
	OBS	0.13 (0.059)	0.11 (0.055)	0.21 (0.064)	0.20 (0.064)
25%	NONE	0.11 (0.061)	0.10 (0.049)	0.18 (0.072)	0.23 (0.055)
	ONE	0.11 (0.062)	0.11 (0.062)	0.18 (0.066)	0.19 (0.063)
	EXP	0.11 (0.064)	0.11 (0.057)	0.18 (0.066)	0.19 (0.062)
	OBS	0.11 (0.062)	0.11 (0.056)	0.18 (0.065)	0.19 (0.063)

^aJ-factor methods used include none (NONE), one for all breeds (ONE), three J-factors calculated using for each breed the expected (EXP), or observed (OBS) breed contribution calculated using BOA from previous analysis [13]
^bStandard deviations in parentheses. Standard deviation can be calculated as \({{{\text{SD}}} \mathord{\left/ {\vphantom {{{\text{SD}}} {\sqrt n }}} \right. \kern-0pt} {\sqrt n }}\), where SD is the standard deviation and n is the number of replicates (100)
^cFor 100% genotyping with ONE, it is equivalent to fitting NONE, EXP, and OBS

ISSN: 1297-9686