A Causality Perspective of Genomic Breed Composition for Composite Animals.

Tom Schultz,Stewart Bauck,Jun He,Yangfan Wang,Guilherme J M Rosa,Ryan Ferretti,John Genho,Xiao-Lin Wu,Zhi Li,Jamie Parham,Richard G Tait

doi:10.3389/fgene.2020.546052

Abstract

Genomic breed composition (GBC) of an individual animal refers to the partition of its genome according to the inheritance from its ancestors or ancestral breeds. For crossbred or composite animals, knowing their GBC is useful to estimate heterosis, to characterize their actual inheritance from foundation breeds, and to make management decisions for crossbreeding programs. Various statistical approaches have been proposed to estimate GBC in animals, but the interpretations of estimates have varied with these methods. In the present study, we proposed a causality interpretation of GBC based on path analysis. We applied this method to estimating GBC in two composite breeds of beef cattle, namely Brangus and Beefmaster. Three SNP panels were used to estimate GBC: a 10K SNP panel consisting of 10,226 common SNPs across three GeneSeek Genomic Profiler (GGP) bovine SNP arrays (GGP 30K, GGP 40K, and GGP 50K), and two subsets (1K and 5K) of uniformly distributed SNPs. The path analysis decomposed the relationships between the ancestors and the composite animals into direct and indirect path effects, and GBC was measured by the relative ratio of the coefficients of direct (D-GBC) and combined (C-GBC) effects from each ancestral breed to the progeny, respectively. Estimated GBC varied only slightly between different genotyping platforms and between the three SNP panels. In the Brangus cattle, because the two ancestral breeds had a very distant relationship, the estimated D-GBC and C-GBC were comparable to each other in the path analysis, and they corresponded roughly to the estimated GBC from the linear regression and the admixture model. In the Beefmaster, however, the strong relationship in allelic frequencies between Hereford and Shorthorn imposed a challenge for the linear regression and the admixture model to estimated GBC reliably. Instead, D-GBC by the path analysis included only direct ancestral effects, and it was robust to bias due to high genomic correlations between reference (ancestral) breeds.

Highlights

Genomic breed composition (GBC) of an individual animal refers to the partition of its genome according to the inheritance from its ancestors or ancestral breeds
We proposed a causality interpretation of genomic breed composition implemented by the path analysis for composite animals in the present study
In Brangus, because the two ancestors breeds are remotely related, or they have a close to zero correlation, the estimated D-GBC agreed with the estimated C-GBC in the path analysis, and they both agreed well with the estimated GBC by the admixture model and linear regression

Summary

Introduction

Genomic breed composition (GBC) of an individual animal refers to the partition of its genome according to the inheritance from its ancestors or ancestral breeds. A haplotype segment is usually present in many individuals as descendants of a common ancestor from which the segment originates. The chance of IBS, is minimal if many markers are included in the segments, e.g., based on runs of homozygosity (ROH), which are long DNA segments containing consecutive homozygous loci (Ferencakovic et al, 2011; Purfield et al, 2012). For crossbred or composite animals, GBC is often used to estimate heterozygosity, to understand their breeding history, to characterize their actual inheritance from foundation breeds, and to make management decisions for crossbreeding programs (VanRaden and Cooper, 2015; Akanno et al, 2017; Gobena et al, 2018; He et al, 2018)

Objectives

Methods

Results

Conclusion