Genomic Prediction Model Based on Haplotype Clusters

• Activity Number: 365
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Computing
• Author(s): Stephen Kachman*
• Companies: University of Nebraska
• Keywords: Bayesian ; GWAS ; Genomic prediction

Abstract

Models used for genomic prediction in large livestock populations are based on the underlying assumption that genetic markers are the casual variants. An alternative model using haplotype clusters as hidden states in a hidden Markov model and genomic effects to be associated with loci located on those unobserved clusters. Genomic effects at the loci are modeled using a prior which is a mixture of a normal and a point mass at zero distributions. The model was implemented as a C++ program and uses openMP to take advantage of multiple CPU cores. The program has been successfully used to construct genomic predictors using a data on 1,152 animals genotyped with mixture of genotyping platforms with a total 100,855 mapped SNP.