Genomic dissection of maternal, additive and non-additive genetic effects for growth and carcass traits in Nile tilapia

Table 4 Corrected estimates of heritability, variance ratios, and variances for the model of best fit for each trait and approach

Traits	Models	HWE				NOIA
Traits	Models	\(\upsigma_{\text{A}}^{2}\)	\(\upsigma_{{{\text{E}}_{{{\mathbf{aa}}}} }}^{2}\)	\(h^{2}\)	\({\mathbf{e}}_{{{\mathbf{aa}}}}^{2}\)	\(\upsigma_{\text{A}}^{2}\)	\(\upsigma_{{{\text{E}}_{{{\mathbf{aa}}}} }}^{2}\)	\(h^{2}\)	\({\mathbf{e}}_{{{\mathbf{aa}}}}^{2}\)
BD	AME	0.086 (0.024)	0.080 (0.049)	0.159 (0.043)	0.147 (0.091)	0.086 (0.024)	0.080 (0.049)	0.159 (0.043)	0.147 (0.091)
BWH	AME	698.774 (267.730)	1169.547 (672.154)	0.099 (0.037)	0.167 (0.096)	698.772 (267.729)	1169.539 (672.149)	0.099 (0.037)	0.167 (0.095)
BL	AM	0.285 (0.107)		0.085 (0.031)		0.284 (0.107)		0.085 (0.031)
FW	AM	117.948 (41.825548)		0.096 (0.0324407)		117.948 (41.825)		0.096 (0.033)
BT	A	1.694 (0.441)		0.174 (0.041)		1.694 (0.441)		0.174 (0.041)
FY	A	1.757 (0.406)		0.191 (0.039)		1.758 (0.406)		0.191 (0.039)

The variances and ratios using the HWE approach were corrected by (mean (leading diagonal) − mean) of the the corresponding relationship matrices as per [9]
The mean of diagonal and off-diagonal elements in relationship matrices using the NOIA approach resulted ~ 1 and 0, respectively. Hence, scaling was not required for the NOIA approach
Standard errors are in parentheses
A model represents the basic animal model, which is gradually expanded to an AME (model with additive (A), maternal (M) and first order additive-by-additive epistatic interactions (E) effects) by adding each effect as random effects
NOIA and HWE are the two approaches used to calculate the different variance components (see “Methods” for details)
The six traits are: BD body depth, BWH body weight at harvest, BL body length, FW fillet weight, BT body thickness, FY fillet yield

ISSN: 1297-9686