Bayesian genomic prediction of junctional epidermolysis bullosa in sheep

Abstract

Bayesian genomic prediction of junctional epidermolysis bullosa in sheep 1 Abstract: Junctional epidermolysis bullosa (JEP) is a heritable skin and mucosa disorders 2 condition in association with mendelian mutations in sheep. The purpose of this investigation 3 is to explore the relationship between different priors, linkage disequilibrium and single 4 nucleotide polymorphisms (SNPs) selection methods to accuracy of Bayesian GP of JEP in 5 sheep. 92 Spanish Churra sheep breed genotyped by 40668 SNP markers. Bayes Cπ shown to 6 have slightly higher predicted accuracy [0.724 (0.113)] by unselected data. Prediction 7 performance of the Bayesian GP models was found to be similar after correction for LD. There 8 was a significant difference between predicted accuracies due to the SNPs selection by ranked 9 p values of whole and training only dataset using linear model. The relevance of genetic 10 architecture in conjugate to the prior distributions clearly supported by the unselected data. The 11 most obvious finding emerge from this study is that preselection of SNPs referring to genetic 12 architecture of the phenotype may lower the needs of computational load. 13