Genetic correlations: a parameter or a latent phenotype in genetic evaluations?

Abstract

Genetic correlations are relevant parameters in genetic evaluations, particularly when a breeding program aims to achieve genetic progress for multiple traits altogether. These correlations are usually estimated from a base population as one of the many parameters that define the distribution used to predict breeding values for the selection candidates. In such a fashion, genetic correlations are assumed to be identical for all selection candidates. However, with a preliminary study on the output predicted breeding values of sires with more than 500 daughters from the French Montbéliarde population, we observed that the genetic correlation among daughters from different sires may differ substantially, i.e., different sires expressed different genetic correlations between traits through their daughters. Thus, if genetic correlations are specific values inherent to each individual, they could be considered as a phenotype; in other words, genetic correlations may be the observable consequence of a concealed regulatory trait guiding the relationship between observable traits. For antagonistic traits (e.g. production and fertility in dairy cattle), it is reasonable to believe that individuals on the extremes of the trade-off distribution are likely to present a low breeding value for this concealed regulatory trait. However, due to our inability to directly measure this potential regulatory trait, it can be considered a latent phenotype. Although a method to consider such hypothesis that genetic correlations may be a latent phenotype is yet undefined, there is no doubt that such hypothesis has an impact on the medium to long-term perspectives of a breeding program, given its breeding goals. Hypothesizing that genetic correlations are latent phenotypes, simulations can then be used to assess the genetic progress for multiple traits of interest in a breeding program over many generations, as well as to assess the trajectory of genetic correlations between traits and the genetic progress of the latent regulatory phenotype driving such correlations. Such comprehension of the genetic progress for the latent phenotype is of particular relevance, since a regulatory trait is likely to impact more than only two antagonistic traits, but many of the traits selected for in a breeding program.