Accuracy of genomic prediction of mean performance of Merino sheep for traits related to reproduction rate

Context The Flock Profile is a genomic test delivered to commercial Merino flocks to provide an estimate of their average Australian Sheep Breeding Values (ASBV) using genomic-only breeding values informed from the Sheep Genetics MERINOSELECT analysis. Aims This study aimed to examine the predictive ability of the Flock Profile genomic test to estimate the mean genetic merit of Merino sheep flocks, estimated through validation using the MERINOSELECT analysis. Methods Data from this analysis were used in a validation study to test the accuracy of predicting mean flock genetic performance for reproductive traits. For each validation flock, the pedigree, genotypes and performance data were removed for the entire flock, and then its Flock Profile result was estimated from genomic predictions based on estimated single-nucleotide polymorphism marker effects from single-step genomic BLUP analyses. The Flock Profile results were then compared with the original ASBVs from the full analysis. Key results The accuracy of the ranking of mean flock performance was high (r > 0.79, median 0.98) for all traits except ewe rearing ability. However, the Flock Profile results were slightly over-dispersed, with an average regression slope across traits of 0.91, and thus had slightly more variation compared with the ASBVs. Genomic predictions for individual animals were also highly correlated (r > 0.78) with the full ASBVs. Conclusions The results from this study have provided the evidence to enable more reproduction traits to be reported in the Flock Profile test. Implications The study also demonstrated variability in genomic connectedness to the reference population across flocks and traits, and thus, the development of an accuracy metric will be important soon. This initial study indicates that further research is needed to evaluate genomic prediction accuracy in less genetically related populations, particularly for flocks lacking phenotypic and pedigree data, to better inform breeding decisions and on-farm management strategies.