Association of genomically enhanced and parent average breeding values with cow performance in Nordic dairy cattle

摘要

This study compared the abilities of virgin heifergenomically enhanced breeding values (GEBV) andparent average breeding values (PA) to predict futurecow performance. To increase confidence in genomictechnology among farmers, a clear demonstration of therelationship between genomic predictions and futurephenotypes is needed. We analyzed 12 different traitsin first parity, including production, conformation, fertility,and other functional traits. Phenotype data wereobtained from national milk recording schemes andbreeding values from the Nordic Cattle Genetic Evaluation.Direct genomic breeding values were calculatedusing genomic BLUP and combined with traditionalbreeding values, using bivariate blending. The datacovered 14,862 Red Dairy Cattle, 17,145 Holstein, and7,330 Jersey genotyped virgin heifers born between2013 and 2015 in Denmark, Finland, and Sweden. Phenotypesadjusted for systematic environmental effectswere used as measures of cow performance. Two methodswere used to compared virgin heifer GEBV andPA regarding their ability to predict future cow performance:(1) correlations between breeding values andadjusted phenotypes, (2) ranking cows into 4 quartilesfor their virgin heifer GEBV or PA, and calculatingactual cow performance for each quartile. We showedthat virgin heifer GEBV predicted cow performancesignificantly better than PA for the vast majority ofanalyzed traits. The correlations with adjusted phenotypeswere 38 to 136% higher for GEBV than for PAin Red Dairy Cattle, 42 to 194% higher for GEBV inHolstein, and 11 to 78% higher for GEBV in Jersey.The relative change between GEBV bottom and topquartiles compared with that between PA bottom andtop quartiles ranged from 9 to 261% for RDC, 42 to138% for Holstein, and 4 to 90% for Jersey. Hence,farmers in Denmark, Finland, and Sweden can haveconfidence in using genomic technology on their herds.