Impact of QTL properties on the accuracy of multi-breed genomic prediction

Table 4 Average estimated heritabilities using different marker densities to calculate the genomic relationship matrix

Sc.	Nb HF ¹	Nb J ²	RANDOM				VAR
			600 k	60 k	600 k + QTL	60 k + QTL	600 k	60 k	600 k + QTL	60 k + QTL
1	2000	2000	0.78 (0.003)	0.77 (0.003)	0.80 (0.002)	0.84 (0.001)	0.59 (0.001)	0.58 (0.001)	0.61 (0.001)	0.64 (0.001)
2	2000	500	0.76 (0.004)	0.74 (0.004)	0.78 (0.003)	0.82 (0.003)	0.54 (0.003)	0.53 (0.003)	0.57 (0.003)	0.59 (0.003)
3	2000	100	0.75 (0.005)	0.73 (0.005)	0.77 (0.004)	0.80 (0.005)	0.54 (0.004)	0.53 (0.004)	0.57 (0.004)	0.59 (0.004)
4	2000	0	0.75 (0.005)	0.73 (0.005)	0.77 (0.004)	0.81 (0.005)	0.55 (0.004)	0.54 (0.004)	0.58 (0.004)	0.60 (0.004)
5	500	2000	0.79 (0.004)	0.78 (0.005)	0.80 (0.003)	0.83 (0.002)	0.61 (0.006)	0.60 (0.005)	0.63 (0.005)	0.66 (0.005)
6	100	2000	0.78 (0.007)	0.77 (0.007)	0.80 (0.006)	0.82 (0.005)	0.58 (0.006)	0.58 (0.006)	0.60 (0.006)	0.63 (0.005)
7	0	2000	0.78 (0.008)	0.77 (0.008)	0.79 (0.007)	0.82 (0.006)	0.56 (0.006)	0.55 (0.006)	0.57 (0.005)	0.60 (0.005)

Average heritabilities (standard errors across replicates) estimated with a model including a random across-breed animal effect and a within-breed animal effect using different reference populations, 100 QTL underlying the trait with on average a moderately low minor allele frequency and using simulated allele substitution effects randomly sampled from a gamma distribution (RANDOM) or with each QTL explaining an equal proportion of the genetic variance (VAR). The genomic relationship matrix was calculated using 606 384 SNPs (600 k), 60 000 SNPs (60 k), 606 384 SNPs plus all sampled QTL (600 k + QTL), or 60 000 SNPs plus all sampled QTL (60 k + QTL).
Sc. = scenarios; ¹Nb HF = number of Holstein Friesian animals; ²Nb J = number of Jersey animals.

ISSN: 1297-9686