# Genetic evaluation of longevity in Australian Angus cattle using random regression models

## TL;DR

This study evaluates cow longevity in Australian Angus cattle using genetic models to improve breeding decisions and beef production sustainability.

## Contribution

The study introduces random regression models to estimate genetic parameters for cow longevity, revealing age-specific heritability patterns.

## Key findings

- Heritability estimates for longevity traits peaked between ages 4 to 6 years.
- Genetic correlations between early and late life longevity measures were low.
- Excluding censored data underestimated breeding values for sires with active daughters.

## Abstract

Cow longevity is an economically important trait for beef breeders directly impacting the profitability and sustainability of beef cattle production systems. Despite its great importance, early selection for longevity is complex because the true longevity of a cow is not known until the end of her life. In this study, we aimed to estimate variance components and genetic parameters for 2 binary measures of cow longevity in Australian Angus cattle population. Traditional longevity (TL) represented the cow’s ability to avoid culling after the first calving while functional longevity (FL) also accounted for calving events while the cow was present in the herd. Five datasets consisting of animals culled for different reasons were created and evaluated separately to compare the estimates of variance components and genetic parameters. We also investigated the impact of censored data on estimated breeding values (EBV) of bulls with different proportions of active daughters. A single-trait random regression model using a Bayesian Gibbs sampler was applied to both longevity traits and all 5 culling reason groups between ages 2 to 11 yr. The heritabilites were generally low and ranged between 0.02 to 0.19 for TL and between 0.02 to 0.20 for FL traits. The peak of heritabilites were found between ages 4 to 6 yr for both longevity measures. The low estimates of genetic correlations between ages at the beginning and end of the trajectory in all culling reason groups indicated that longevity evaluated at early and late stages of life are not genetically the same traits. The EBV of sires with active daughters were underestimated when the censored data was excluded from the analysis. The negative impact of censoring was larger for younger sires who had a larger proportion of active daughters. Our results indicate the additive genetic component has a sizeable contribution to the variability of longevity in Australian Angus cattle and therefore, the genetic improvement of longevity can be achieved if longevity is considered as a long-term breeding objective.

Cow longevity is an economically important trait for Australian beef breeders that requires further genetic improvement. Random regression models can enhance the genetic evaluations of longevity if longevity is considered a long-term breeding goal.