Fig. 4: The genomic prediction accuracy of the Bayesian and BLUP alphabets for different combinations of sample size, marker density, heritability, and QTL size (5 and 20% markers as causal variants).
From: Performance of Bayesian and BLUP alphabets for genomic prediction: analysis, comparison and results
It can be seen that with an increase in heritability, genomic prediction accuracy also increased irrespective of sample size, marker density, and QTL size. The accuracy is also increased with an increase in the sample size and the marker density. The genomic prediction accuracies are seen to be lowest for SBLUP, irrespective of the trait genetic architecture. The GBLUP secured higher accuracy for the traits controlled by many QTLs, each having a small effect. On the other hand, the Bayesian methods produced higher accuracy when a few QTLs governed the trait, each with a larger effect on the genotypic variability.
