A bivariate genomic model with additive, dominance and inbreeding depression effects for sire line and three-way crossbred pigs

Ole F Christensen,Anders B Strathe,Guosheng Su,Per Madsen,Bjarne Nielsen,Tage Ostersen,Ingela Velander,Tao Xiang

doi:10.1186/s12711-019-0486-2

Abstract

BackgroundCrossbreeding is widely used in pig production because of the benefits of heterosis effects and breed complementarity. Commonly, sire lines are bred for traits such as feed efficiency, growth and meat content, whereas maternal lines are also bred for reproduction and longevity traits, and the resulting three-way crossbred pigs are used for production of meat. The most important genetic basis for heterosis is dominance effects, e.g. removal of inbreeding depression. The aims of this study were to (1) present a modification of a previously developed model with additive, dominance and inbreeding depression genetic effects for analysis of data from a purebred sire line and three-way crossbred pigs; (2) based on this model, present equations for additive genetic variances, additive genetic covariance, and estimated breeding values (EBV) with associated accuracies for purebred and crossbred performances; (3) use the model to analyse four production traits, i.e. ultra-sound recorded backfat thickness (BF), conformation score (CONF), average daily gain (ADG), and feed conversion ratio (FCR), recorded on Danbred Duroc and Danbred Duroc-Landrace–Yorkshire crossbred pigs reared in the same environment; and (4) obtain estimates of genetic parameters, additive genetic correlations between purebred and crossbred performances, and EBV with associated accuracies for purebred and crossbred performances for this data set.ResultsAdditive genetic correlations (with associated standard errors) between purebred and crossbred performances were equal to 0.96 (0.07), 0.83 (0.16), 0.75 (0.17), and 0.87 (0.18) for BF, CONF, ADG, and FCR, respectively. For BF, ADG, and FCR, the additive genetic variance was smaller for purebred performance than for crossbred performance, but for CONF the reverse was observed. EBV on Duroc boars were more accurate for purebred performance than for crossbred performance for BF, CONF and FCR, but not for ADG.ConclusionsMethodological developments led to equations for genetic (co)variances and EBV with associated accuracies for purebred and crossbred performances in a three-way crossbreeding system. As illustrated by the data analysis, these equations may be useful for implementation of genomic selection in this system.

Highlights

Crossbreeding is widely used in pig production because of the benefits of heterosis effects and breed complementarity
The aims of this study were to (1) present a modification of the model of Vitezica et al [12] and Xiang et al [13] for the case of a purebred sire line and three-way crossbred pigs; (2) based on this model, present equations for genetic variances, genetic covariance, and estimated breeding values (EBV) for purebred and crossbred performance; (3) use the model to analyse four production traits, i.e. ultra-sound recorded backfat thickness, conformation score, average daily gain, and feed conversion ratio, recorded on Danbred Duroc and Danbred DLY pigs raised in the same environment; and (4) obtain estimates of genetic parameters, genetic correlations between purebred and crossbred performances, and EBV with associated accuracies for purebred and crossbred performances for this data set
BF ultra-sound recorded backfat thickness, CONF overall conformation score, ADG average daily gain, FCR feed conversion ratio removed from the model

Summary

Introduction

Crossbreeding is widely used in pig production because of the benefits of heterosis effects and breed complementarity. There may be systematic differences in the environments of, for example, nucleus herds and finisher herds For these reasons, selecting animals for high purebred performance may not result in selecting those that lead to high-performance crossbred offspring in a specific crossbreeding scheme. An approach to handle this is to define two breeding values for purebred animals, the usual breeding value for purebred performance (mating within breed) and an alternative breeding value for crossbred performance (mating with the other breed), as in Wei and van der Werf [2], and select based on an index of the two estimated breeding values (EBV) [3, 4] These two breeding values are correlated, with the correlation depending on the size of non-additive genetic effects in combination with differences in allele frequencies between lines, and possibly in addition, on genotype by environment interactions

Objectives

Methods

Results

Discussion

Conclusion