Summary and Implications
We propose a novel model (BayesN) for genomic prediction, where multiple markers in a small segment are simultaneously fitted to jointly capture the effect of major genes (QTL) in the segment. Compared with BayesB, in which the effects of neighboring markers are a prioriassumed to be independent, BayesN gave higher accuracies of prediction and required less computing effort. BayesN is an accurate and practical method for analyzing high-density markers, especially for traits influenced by rare QTL alleles
Iowa State University
Zeng, Jian; Garrick, Dorian J.; Dekkers, Jack C.; and Fernando, Rohan L.
"A Nested Mixture Model for Genomic Prediction Using Whole-Genome SNP Genotypes,"
Animal Industry Report:
AS 662, ASL R3060.
Available at: https://lib.dr.iastate.edu/ans_air/vol662/iss1/21