| Popis: |
A new tract scoring system for age at puberty (RTSCORE) has been proposed for application in Northern Australian beef herds. RTSCORE is an indirect measure of age at first corpus luteum (AGECL), which is an accurate, intensively collected phenotype but has limited feasibility for large scale commercial application. When measurements are taken in unrelated herds it is possible to use multivariate genomic prediction to compare RTSCORE to other puberty related traits, such as AGECL. Here we use this methodology to evaluate the ability of RTSCORE to accurately predict genomic estimated breeding values (GEBV) for age at puberty. We also evaluate the effect of including AGECL phenotypes on GEBV accuracy for RTSCORE. Multi-trait best linear unbiased prediction with a multi-breed, across-population genomic relationship matrix derived from 728,000 imputed SNP markers was used to compare AGECL in a population of 894 Brahman and 1088 Tropical Composite heifers and RTSCORE in a population of 979 Brahman, 1803 Santa Gertrudis, and 914 Droughtmaster heifers. GEBV accuracy from within-herd predictions of puberty were compared to GEBV accuracies from a combined reference population containing both measures of puberty. The inclusion of heifers with RTSCORE and corresponding genotypes improved the accuracy of genomic evaluations for AGECL compared to within-herd predictions. Although across-herd prediction of puberty using RTSCORE for training resulted in accuracies that were low, this was expected given a correlation between RTSCORE and AGECL of -0.66. These results suggest that RTSCORE is an appropriate measure of puberty and can be used to accurately predict age at puberty. Including both populations in the reference herd for prediction of RTSCORE resulted in either no change or slight improvement in accuracy. As RTSCORE is not an exact measure of puberty, AGECL should be included with RTSCORE when predicting genomic breeding values for puberty in a commercial setting. |
