Disease & heat resistance

Genetic Assessment of Resilience Indicators Derived from Dry Matter Intake, and Feeding Behavior Using Random Regression Models in Beef Cattle

---

Abstract

Resilience to environmental challenges is a critical determinant of productivity and welfare in tropical beef production systems. Using high-frequency feeding behavior data collected through precision livestock technologies provides a robust and scalable approach to identifying resilient cattle. Therefore, this study aimed to derive resilience indicators from predictive models of daily dry matter intake (DMI), feeding time (FT), and feeding events (FE), measured using electronic feeders, and to estimate heritability (h²) and genetic correlations (rg) using genomic information for these resilience indicators, feeding behavior traits, average daily gain (ADG) and yearling weight adjusted to 378 days of age (W378). Data were obtained from the Institute of Animal Science, SP (Brazil). The phenotypic dataset comprised records collected between 2013 and 2024 from 1,353 growing Nellore cattle and 77,784 individual feeding-behavior measurements obtained using GrowSafe® electronic feeders. Longitudinal trajectories of DMI, FT, and FE were characterized using random regression models (RRM) with third-order Legendre orthogonal polynomials for DMI and fourth-order polynomials for FT and FE. Residuals from the RRM were used to derive two resilience indicators for each trait: (1) the natural logarithm of residual variance (DMILnVar, FTLnVar, FELnVar); (2) first-order autocorrelation (DMIrauto, FTrauto, FErauto). For these indicators, values closer to zero indicate greater resilience. The pedigree included 13,351 individuals, and the genomic dataset comprised 3,860 genotyped animals with 383,856 SNPs. The (co)variance components were estimated in a ssGBLUP framkeword considering the AIREML algorithm in the blupf90+ software. The h² estimates were: 0.26 ± 0.04 (DMI), 0.24 ± 0.04 (DMILnVar), 0.03 ± 0.03 (DMIrauto), 0.42 ± 0.05 (FT), 0.34 ± 0.05 (FTLnVar), 0.01 ± 0.02 (FTrauto), 0.35 ± 0.05 (FE), 0.16 ± 0.04 (FELnVar), 0.02 ± 0.03 (FErauto), 0.36 ± 0.04 (W378), and 0.28 ± 0.04 (ADG). The genetic correlations for DMILnVar were high and positive, with estimates of 0.65 ± 0.11 (DMI), 0.74 ± 0.11 (W378), and 0.64 ± 0.14 (ADG). For DMIrauto, the correlation estimates were 0,18 ± 0.54 (DMI), 0.19 ± 0.53 (W378), and 0.09 ± 0.46 (ADG). Correlations between FTLnVar and DMI, P378, and ADG were moderate and negative, ranging from -0.06 ± 0.17 to -0.17 ± 0.13. Estimates close to zero were found between FTrauto and the performance traits. The results indicate genetic variability for the resilience indicators studied, which can be useful for genomic selection purposes. This study is the first to investigate relationships between performance and resilience indicators derived from longitudinal feed intake and feeding behavior data in beef cattle. The findings suggest a strong connection between resilience indicators derived from feeding behavior and performance traits, showing that more resilient individuals may not be those with higher performance for ADG and W378.

Keywords: 2026

How to Cite:

Araújo, L., Rodrigues, G., Cyrillo, J., Valente, J., Feitosa, F., Canesin, R., Bonilha, S. & Mercadante, M., (2026) “Genetic Assessment of Resilience Indicators Derived from Dry Matter Intake, and Feeding Behavior Using Random Regression Models in Beef Cattle”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2295736. doi: https://doi.org/10.31274/wcgalp.24367

Rights: 1