Disease & heat resistance

How may selection on statistical resilience indicators affect biological mechanisms and realised performance and health? A simulation study from a mechanistic model of parasite infections in sheep

---

Abstract

Disease resilience, the ability of an animal to be minimally affected by an infectious challenge or recover quickly, is a desirable trait in animal breeding. Resilience involves multiple biological mechanisms and cannot be measured directly. Recently, statistical resilience indicators (RIs) based on longitudinal performance data of animals have been proposed as novel resilience phenotypes. Although RIs lack biological support, they are shown to be heritable. Reported genetic correlations with production, health and fitness traits vary from not significant to strongly favourable or unfavourable across contexts. This raises questions about when genomic selection using RIs will improve productivity and health, and which biological mechanisms are affected by selecting for these RIs. Here, we used stochastic simulation to study the genetic relationship between an RI (area under the curve between expected and realised growth) and observable performance and health traits (realised growth, faecal egg count (FEC)), as well as underlying biological traits (growth potential, immunocompetence, and anorexia) in a sheep population challenged by gastrointestinal parasite. We simulated genetic latent traits such as growth potential (growth in the absence of challenge), immunocompetence (ability to inhibit larvae establishment, survival and fecundity of worms) and degree of anorexia (infection-induced reduction in voluntary feed intake). Each sheep is assumed to be infected. The within-host infection dynamics were simulated using an established mechanistic resource allocation model that partitions ingested nutrients into maintenance, growth and immune functions predicting realised growth and health. Expected growth is simulated assuming no infection. Bivariate REML analyses were used to calculate the genetic correlation between RI, FEC, and production (growth potential and realised growth). We assumed three scenarios for the genetic correlation between the immunocompetence and growth parameters as zero, moderately antagonistic and moderately synergistic. Each study was replicated 100 times with a population containing 270 sires, 10 dams per sire and 3 offspring per mating, yielding small confidence intervals for estimated genetic correlations and heritabilities (< 0.05). Our results show RI has a strong favourable genetic correlation with immunocompetence and FEC, independent of the assumed correlations between underlying traits. However, an unfavourable negative genetic correlation was observed between the RI and growth potential, suggesting that selection for resilience may reduce growth in ideal conditions. The genetic relationship between RI and realised growth under infection depends on anorexia level: when anorexia is low, the genetic correlation is negative, whereas it becomes positive as anorexia increases. Nevertheless, the trade-off between health and realised growth is much weaker than between RI and realised growth, suggesting selection on direct health may be more beneficial. Our results highlight the need for caution when incorporating RIs into multi-trait breeding objectives and, where possible, combining them with health and production records in genetic evaluations to balance health and productivity.

Keywords: 2026

How to Cite:

Ghaderi Zefreh, M., Pong-Wong, R. & Doeschl-Wilson, A., (2026) “How may selection on statistical resilience indicators affect biological mechanisms and realised performance and health? A simulation study from a mechanistic model of parasite infections in sheep”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2285676. doi: https://doi.org/10.31274/wcgalp.23775

Rights: 1