Skip to main content

Table 4 Accuracya of the DANZ training predictions for Fat, Milk and Protein Yield in the Red Holstein bull and the Australian Red cow validation sets

From: Exploiting biological priors and sequence variants enhances QTL discovery and genomic prediction of complex traits

 

FAT

MILK

PROTEIN

Analytical Modelb

Red Hol

Aust Red

Red Hol

Aust Red

Red Hol

Aust Red

BayesR 800 K

0.565 (0.001)

0.344 (0.003)

0.650 (0.001)

0.317 (0.003)

0.603 (0.001)

0.200 (0.001)

BayesR SEQ

0.572 (0.001)

0.354 (0.004)

0.663 (0.002)

0.308 (0.005)

0.612 (0.001)

0.220 (0.003)

BayesRC Lact

0.576 (0.002)

0.353 (0.002)

0.664 (0.001)

0.325 (0.004)

0.616 (0.001)

0.226 (0.003)

BayesRC RLact

0.571 (0.001)

0.352 (0.002)

0.657 (0.001)

0.302 (0.005)

0.612 (0.001)

0.218 (0.002)

  1. aEstimated as the average correlation between the genomic prediction and corrected phenotypes. The highest accuracy is in bold font in each column. Numbers in in brackets indicate relative convergence of 5 independent Bayesian MCMC chains (estimated from [SD of the mean accuracy]/√5). Note: the numbers in brackets should not be interpreted as a “standard error” because they are estimated from 5 Bayesian MCMC chains run on the same data set
  2. bBayesR models used either 800 K SNP array (600,640 genotypes) or 994,019 sequence variants (SEQ). The BayesRC model definitions are given in Table 2
Back to article page

BMC Genomics

ISSN: 1471-2164

Contact us