Phenomics

Genetic parameters for sensor-based behavior traits and their relationship with milk production in Holstein- Friesian dairy cattle

---

Abstract

The objective of this research was to estimate genetic parameters and identify the potential gain from adding behavioral traits as proxies for the genetic evaluation of milk production (MY, kg/day). Behavioral traits, including rumination time (RT, min/day) and daily activity (DA, min/day), were collected from Holstein-Friesian cows in three commercial dairy herds equipped with a sensor-based system. The research considers daily results from 6,514 lactations by 4,599 cows milked between 2019 and 2023, located in the central and central-eastern regions of Mexico. Each herd was managed independently regarding feeding and diet formulations. Data preparation and estimation of descriptive statistics were performed using SAS (version 9.4). Available information was collected through the Allflex DataFlow™ II system, and RT and DA measurements were obtained from the SCR Heatime® HR System (Copyright© 2015 SCR Engineers Ltd). Records of cows with missing data on RT or MY were excluded. Finally, contemporary groups were defined by combining herd, year, and season (HYS) information according to the date of data recording. Contemporary groups with fewer than 5 animals were merged with adjacent groups. The (co)variance components were estimated using uni- and bivariate animal models for the MY, RT, and DA traits, with the restricted maximum likelihood method and the AI-REML algorithm implemented in the BLUPF90+ program. The effects included in the models used were fixed effects: the contemporary group of herd-year-season (HASj) and parity (NPk); and additive genetic (ai), permanent environment (pei), and residual random effects (eijkl). The results obtained showed heritability (h2) values of 0.18±0.02, 0.47±0.06, and 0.11±0.03 and repeatability values of 0.18±0.02, 0.54±0.01, and 0.58±0.01 for MY, RT, and DA, respectively. Genetic (𝑟̂𝑎) and phenotypic (𝑟̂𝑓) correlations between RT-MY were 0.19±0.06 and 0.11±0.02, between RT-DA of -0.05±0.15 and -0.06±0.03, and between DA-MY of 0.43±0.14 and 0.21±0.02, respectively. MY, RT, and DA showed no genetic, phenotypic, or environmental trends across sire birth years (2016-2019). RT presented moderate heritability, high repeatability, and low genetic correlation with MY. At the same time, DA showed low heritability and high repeatability, as well as null genetic and environmental associations with RT, but a moderate-low correlation with MY. These results suggest that RT may reflect increased productive performance, related to better resource use per unit of milk, i.e., a distinction between energy use for productive processes and metabolic processes in the cow. This research provides evidence for incorporating behavioral phenotypes into selection programs for milk production in Holstein cattle. Further validation across genetically diverse populations, combined with integration of genomic data, would enhance prediction accuracy and extend the practical applicability of these behavioral selection tools in commercial dairy breeding programs.

Keywords: 2026

How to Cite:

González-Sandoval, A., Ruíz-López, F., García-Ruíz, A., Romano-Muñoz, J. & Toledo-Alvarado, H., (2026) “Genetic parameters for sensor-based behavior traits and their relationship with milk production in Holstein- Friesian dairy cattle”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286919. doi: https://doi.org/10.31274/wcgalp.24165

Rights: 1