Vector space algebra for scaling and centering relationship matrices under non-Hardy–Weinberg equilibrium conditions

Table 3 Estimates of variance components for viral load for the crossbred dataset, using genomic relationship matrices constructed with alternative biological parameterizations: Hardy–Weinberg equilibrium (HW), non-Hardy–Weinberg equilibrium (NO-HW), natural and orthogonal interactions approach (NOIA) and Gram- Schmidt process for additive (GSP-A), dominance (GSP-D) and orthonormal additive (GSP-N)

Parameters	HW	NO-HW	NOIA	GSP-A	GSP-D	GSP-N
\(h^{2}\)	4.30	4.01	16.08	16.16	0.00	15.86
\(d^{2}\)	20.35	20.44	1.79	1.80	25.11	0.00
\(V_{A}\)	58.68	54.63	220.47	221.82	0.0032	217.10
\(V_{D}\)	277.90	278.31	24.57	24.71	340.86	0.000
\(V_{RESIDUAL}\)	1029.28	1028.70	1125.84	1125.84	1016.80	1152.07
\(V_{PHENOTYPIC}\)	1365.86	1361.64	1370.88	1372.37	1357.66	1369.17

\(h^{2}\): heritability × 100; \(d^{2}\): dominance × 100; heritability and dominance are the additive or dominance variance divided by the sum of all variance components; \(V_{A}\): Additive variance; \(V_{D}\): dominance variance; \(V_{RESIDUAL}\) residual variance; \(V_{PHENOTYPIC} = V_{A} + V_{D} + V_{RESIDUAL}\)

ISSN: 1297-9686