Estimation & Prediction

Causal Architecture of Functional Longevity in Limousine Cattle

---

Abstract

Longevity is a complex, time-dependent trait that is difficult to improve due to its multifactorial nature. Many efforts are currently underway to boost prediction of survival using, improved phenotyping and the use of multi-omics. Here, we explore a complementary route, based on the premise that "true longevity" in cattle is often obscured by numerous confounding factors. We compared a "baseline" multi trait selection model with a structural model that explicitly assumes longevity to be influenced by multiple underlying phenotypes, grouped into (i) morphological traits (e.g., conformation type traits), (ii) fertility traits (age at first calving and first calving interval), and (iii) performance-related traits (weaning and yearling weights). Longevity was assessed through stayability (STAY), which represents the likelihood of a cow surviving and remaining productive in the herd up to a specified calving or age. Calving records included cows born between 1977 and 2023. The final dataset comprised 38,188 cows, daughters of 6,194 sires belonging to 2,471 herds located across the entire national territory. We compared the change of genetic correlations among morphological, fertility and performance traits between the baseline and structural models and further quantified the inferred causal effects among traits. The largest change in genetic correlations was observed for early stayability traits, the same was observed for the "casual" effect of different phenotypes on longevity. The estimated causal effects on longevity were positive across all trait groups, with the strongest influence arising from fertility traits particularly the first calving interval. Regarding genetic correlations, we observed that the causal model generally reduced the magnitude of correlations among traits, suggesting a clearer separation of direct and indirect effects. In particular, correlations between fertility and longevity decreased once causal pathways were accounted for, indicating that part of their apparent association in the baseline model was mediated through intermediate traits. In conclusion, our findings indicate that (i) the inferred causal effects of phenotypes are more pronounced for early stayability classes; and (ii) these results highlight the need to re-evaluate current selection schemes by explicitly considering causal relationships among traits rather than relying solely on correlations.

Keywords: 2026

How to Cite:

Callegaro, S., Mancin, E., Maltecca, C., Bozzi, R. & Tiezzi, F., (2026) “Causal Architecture of Functional Longevity in Limousine Cattle”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2285673. doi: https://doi.org/10.31274/wcgalp.23774

Rights: 1