Genotype × environment interactions in pig breeding programmes. I. Central test

Abstract

In this first paper of a series, Dutch central test results are examined for genotype × environment interaction (G × E) and the data are further used to estimate genetic correlations between the various evaluations of growth and carcass quality, as used in the distinct environments of the breeding programme. G × E in pig breeding programmes is outlined as genotypes expressing different phenotypes in the distinct levels of the breeding programme or even in different environments within a level (e.g. herds). In most studies on G × E, the expectation of genetic correlation between similar traits measured in different environments has been taken to be one. Estimated correlations between similar traits measured on central tested pigs in this study indicate however, that expectations should be smaller, especially for carcass characteristics. Genetic correlations of carcass backfat thickness (CB) with ultrasonic backfat thickness (UB), normally used in on-farm tests, were 0.61 and 0.57 for Dutch Landrace (NL) and Dutch Yorkshire (GY), respectively. Correlations of UB with backfat class, as used in commercial fattening, were 0.25 and 0.42. Genetic correlations of ham + loin % with type class were 0.60 and 0.94. In future analyses of G × E these differences in genetic background of the traits should be taken into account. Genotype × batch and genotype × sex interactions were investigated for daily gain and feed conversion ratio. No significant interactions were found. However, for daily gain between arriving at the station and the end of the test, as well as for weight for age at the end of the test, genotype × batch interaction was significant ( P < 0.05). The possible causes of these interactions are discussed. For slaughter characteristics genotype × month interactions were not of significance.