Carcass trait evaluations – phase II

The project was completed by SRUC and the results summarised as follows:

At the start of the project over 3.9 million carcass records from 2001 to 2014 were available from six different abattoirs. At the completion of the project, seven abattoirs are supplying data – in some cases records are transferred routinely (i.e. weekly) via an automated email and data upload system – with 4.8 million carcass records available.

– It is thought that these data represents about 31% of the national slaughter population, this varies across years but is higher in recent years where we have more data provided.

– Systems to clean, validate and consolidate data have been put in place to maximise the available data.

– Investigations were undertaken to develop appropriate statistical models, in particular methods of accounting for hybrid vigour given the cross bred nature of the carcass data.

– Genetic parameters have been estimated for five carcass traits. In all cases the traits were found to be heritable. The moderate to high heritability’s estimated show that the traits are highly suitable for genetic selection. Heritability estimates for net weight, conformation, fat, and age at slaughter were 0.40, 0.41, 0.45, and 0.63, respectively.

– EBVs were produced for nearly 3 million UK beef and dairy animals.

– Comparing the EBVs across different breeds show that while there are some small differences in ranking, in general there is much more variation within a breed than across the breeds. This demonstrates that the most progress will be made by selecting animals with favourable EBVs within the breed you currently have than by substituting breeds, as no one breed is better than another.

– Genetic trends show little change in recent years. This is not surprising given that until now the tools did not exist to directly select for these carcass traits. Using proxy traits the pedigree sector have made significant genetic improvements but with a poor penetration rate and uptake of performance recorded bulls in the commercial sector the impact has not been observed.

– Now that EBVs are available for the traits of direct interest to the commercial sector clear market signals can occur and hopefully will stimulate commercial finishers to source bulls with favourable carcass trait genetics and we will start to see an improvement in the commercial animals with more animals having the right genetics to better meet and respond to market specifications.

– In particular, age at slaughter has been identified as a very valuable trait to select for as this has a huge impact on the profitability of the beef enterprises.

The Problem:

By utilising existing data from industry and government databases, a previous feasibility project has demonstrated that abattoir data, when salvaged and combined with other national databases, are suitable for producing national beef genetic evaluations that are an improvement on the existing evaluations, where correlated live measurements (liveweight, ultrasound measures) on performance recorded individuals are used. This project is a follow on from that feasibility study and was focused on producing genetic parameters and genetic evaluations for carcass weight, conformation and fat class for those breeds represented with sufficient data in the dataset.

Aims and Objectives:

The aim of this project, following the feasibility project, was to gather more abattoir data from a wider range of abattoirs and make the data retrieval process automatic for each abattoir. The subsequent use of these data in this project has produced genetic evaluations for carcass traits. Results will be disseminated to users by loading into the BASCO database.

The specific objectives were:

1. Estimate genetic parameters, develop models and estimate breeding values for carcass traits (fat, weight and conformation).
2. Calculate correlations between these new EBVs and other important traits and then calculate index weights to allow the new EBVs to be incorporated into existing overall indexes.

Approach:

Data produced in the feasibility study were updated and extracted. They were then used to estimate genetic parameters such as heritability for individual carcass traits and genetic and phenotypic correlations between carcass traits of interest.  Appropriate genetic and statistical models were investigated to improve the accuracy of the EBVs and BLUP analyses performed to produce EBVs.

Objective 1 focused on a within breed approach to the evaluation of the carcass traits. As the carcass data are across all breeds in the UK (both numerous and less so) an across breed evaluation proved possible and highly beneficial to all beef animals (pure and crossbred). Also, the less numerous breeds have sufficient data to produce sufficiently accurate EBVs, where this may not have been feasible in a within breed setting. The first step to implementation of the results is to liaise with the non BASCO breeds to fill in the pedigree gaps and enrich the pedigree to fully utilise an across breed dataset.

Correlations and index weights for the 3 new EBVs have been calculated for their incorporation into breeding goals.