Disease & heat resistance

Estimation of genetic parameters of resilience indicators from feeding behavior in the Iberian pig

---

Abstract

Animal resilience refers to the capacity of an animal to maintain stable performance under environmental or health-related challenges and to recover rapidly to its original or optimal state. Despite its growing relevance in breeding programs, there is still no consensus on the most appropriate indicator to quantify resilience. The objective of this study was to evaluate the genetic potential of three resilience indicators derived from feeding behavior: (a) the natural logarithm of the mean squared error (lnMSE), (b) skewness (Skew), and (c) first-order autocorrelation (Lag1). For each animal, a simple linear regression of the trait was fitted. The residuals, representing deviations from expected performance, were used to compute lnMSE, Skew, and Lag1. These indicators capture complementary dimensions of resilience: lnMSE reflects the overall magnitude of deviations, Skew identifies asymmetric perturbations, and Lag1 quantifies the temporal persistence of deviations. Indicators were computed from records of daily feed intake (DFI), time per visit to the feeder (TPV), and number of visits per day (NVD) from 1,573 purebred Iberian males monitored between 80 and 180 days of age at a commercial testing station in Spain (2020-2025). The pedigree included 2,899 animals. Variance components, heritabilities, and genetic correlations were estimated using bivariate animal models through Bayesian analysis with a Gibbs sampler. Heritability estimates for lnMSE were 0.305, 0.334, and 0.233 for DFI, TPV, and NVD, respectively, and were higher than those obtained for Skew and Lag1. For Skew, heritabilities were 0.167, 0.082, and 0.161, while for Lag1 they were 0.175, 0.095, and 0.147. Genetic correlations showed heterogeneous and in some cases antagonistic patterns: Lag1 indicators generally showed positive correlations with each other, lnMSE exhibited more neutral to slightly negative correlations, while Skew displayed the most divergent correlations (0.538 to −0.836). Some were clearly positive ( >0.70), such as lnMSE for TPV and Skew for NVD (0.912) or Skew for TPV and lnMSE for NVD (0.798). In contrast, others were strongly negative, such as Skew and lnMSE for TPV (-0.871) or Skew for DFI and Lag1 for TPV (−0.820). Given the low heritabilities and moderate sample size, these correlations should be interpreted with caution. Overall, the results indicate that these indicators exhibit heritabilities high enough to support breeding strategies aimed at improving resilience. However, the variability in genetic correlations suggests that each indicator captures distinct aspects, highlighting the need to refine or combine metrics for a more comprehensive characterization of resilience. The antagonistic relationships between some trait pairs may represent genuine opposing genetic mechanisms but could also be influenced by statistical uncertainty due to low heritabilities, underscoring the need for larger datasets and further validation before drawing definitive conclusions about the genetic architecture of resilience traits or making recommendations for indicator selection in breeding programs.

Keywords: 2026

How to Cite:

Barceló-Blasco, F., Nuñez, P., Ibáñez-Escriche, N., Casellas, J., Sánchez-Díaz, M., Negro, S. & Varona, L., (2026) “Estimation of genetic parameters of resilience indicators from feeding behavior in the Iberian pig”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2285409. doi: https://doi.org/10.31274/wcgalp.23689

Rights: 1