Skip to main content

Table 8 Correlation between genomic predictions and deregressed proofs and reliability of genomic predictions for each trait, based on the real 54K dataset

From: Genomic relationships based on X chromosome markers and accuracy of genomic predictions with and without X chromosome markers

Traits Correlation Reliability %
G(A) G(A) + G(X) Gc(A + X) + Pol G(A) G(A) + G(X) Gc(A + X) + Pol
Milk 0.674a 0.676a 0.681b 48.7 48.9 49.6
Fat 0.663a 0.667ab 0.670b 47.1 47.6 48.0
Protein 0.655a 0.657a 0.666b 45.9 46.2 47.5
Growth 0.665a 0.665a 0.668a 47.2 47.2 47.6
Fertility 0.520a 0.532b 0.538b 40.7 42.6 43.5
Birth index 0.517a 0.518a 0.518a 32.5 32.7 32.7
Calving index 0.452a 0.454a 0.452a 30.3 30.5 30.2
Udder health 0.563a 0.568a 0.569a 39.5 40.1 40.3
Other diseases 0.447a 0.459b 0.481c 36.3 38.2 41.9
Body conformation 0.480a 0.478a 0.480a 27.6 27.4 27.6
Feet and legs 0.452a 0.456a 0.457a 33.2 33.7 33.9
Udder conformation 0.595a 0.598a 0.598a 44.0 44.5 44.4
Milking ability 0.642a 0.644a 0.644a 47.1 47.4 47.3
Temperament 0.342a 0.342a 0.348a 18.3 18.3 19.0
Longevity 0.463a 0.468b 0.494c 31.1 31.8 35.4
Average 0.542 0.545 0.551 38.0 38.5 39.3
  1. G(A): model with a G matrix built with autosomal markers only; G(A) + G(X): model with an autosome G matrix and an X chromosome G matrix; Gc(A + X) + Pol: model with a G matrix built with all markers plus a residual polygenic effect; a,b,cCorrelations within a trait without common superscript differed significantly (P < 0.05), according to Hotelling-Williams’ t-test.