Estimation & Prediction

Modelling non-genetic correlations between animal- and litter-recorded traits affect genetic covariances

---

Abstract

In standard multi-trait models, non-genetic covariances between animal- and litter-recorded traits are often ignored. This study investigated the hypothesis that including non-genetic covariances between animal- and litter-recorded traits in a multi-trait model changes the estimates of genetic variances and covariances. The non-genetic covariances covered covariances between residual effects for litter-recorded traits and common birth litter effects for animal-recorded traits, as well as covariances between residual effects for animal-recorded traits and permanent sow effects for litter-recorded traits. Using large multi-trait models for DanBred Landrace pigs, we estimated variances and covariances both with and without accounting for non-genetic covariances. The model included the animal-recorded traits: piglet and weaner growth, finisher growth, lean meat percentage, feed conversion ratio, and conformation score, as well as the litter-recorded traits: litter size, and piglet survival. The results showed that almost all estimated non-genetic covariances between animal-recorded and litter traits were significantly different from zero. Furthermore, incorporating the non-genetic covariances influenced several of the genetic covariances. Particularly the covariances among traits measured around farrowing, i.e. piglet and weaner growth, and litter size. The most significant change in genetic covariance was between litter size and piglet and weaner growth, which changed from negative to positive. Additionally, some genetic variances were notably influenced by the different models. These findings suggest that multi-trait models combining litter- and animal-recorded traits should include non-genetic litter covariances and non-genetic animal covariances to accurately separate genetic and non-genetic (co)variances, leading to more precise estimates.

Keywords: 2026

How to Cite:

Jensen, L., Christensen, O. & Ostersen, T., (2026) “Modelling non-genetic correlations between animal- and litter-recorded traits affect genetic covariances”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2284514. doi: https://doi.org/10.31274/wcgalp.23572

Rights: 1