Abstract #375
Section: Breeding and Genetics
Session: Breeding and Genetics II: Health
Format: Oral
Day/Time: Tuesday 3:00 PM–3:15 PM
Location: 324
Session: Breeding and Genetics II: Health
Format: Oral
Day/Time: Tuesday 3:00 PM–3:15 PM
Location: 324
# 375
Single-step genomic evaluation of digital dermatitis in Canadian Holsteins.
F. Malchiodi*1, D. A. L. Lourenco2, I. Misztal2, A.-M. Christen3, J. Jamrozik1,4, F. S. Schenkel1, D. F. Kelton5, F. Miglior1,4, 1Centre for Genetic Improvement of Livestock, University of Guelph, Guelph, ON, Canada, 2Department of Animal and Dairy Science, University of Georgia, Athens, GA, 3Valacta, Sainte-Anne-De-Bellevue, QC, Canada, 4Canadian Dairy Network, Guelph, ON, Canada, 5Department of Population Medicine, Ontario Veterinary College, University of Guelph, Guelph, ON, Canada.
Key Words: digital dermatitis, single-step
Single-step genomic evaluation of digital dermatitis in Canadian Holsteins.
F. Malchiodi*1, D. A. L. Lourenco2, I. Misztal2, A.-M. Christen3, J. Jamrozik1,4, F. S. Schenkel1, D. F. Kelton5, F. Miglior1,4, 1Centre for Genetic Improvement of Livestock, University of Guelph, Guelph, ON, Canada, 2Department of Animal and Dairy Science, University of Georgia, Athens, GA, 3Valacta, Sainte-Anne-De-Bellevue, QC, Canada, 4Canadian Dairy Network, Guelph, ON, Canada, 5Department of Population Medicine, Ontario Veterinary College, University of Guelph, Guelph, ON, Canada.
This study aimed to assess the gains in the breeding value reliability using single step genomic evaluation for digital dermatitis. Hoof lesions were recorded by 51 hoof trimmers during routine trimming visits in 1,080 herds located in Alberta, British Columbia, New Brunswick, Ontario, and Quebec between 2009 and 2016. The final data set contained 249,801 observations from 105,557 cows. The final pedigree file contained 351,215 animals. Of those animals 10,924 (8,934 bulls and 1,990 cows) also had genomic information and 3,637 cows had both phenotypic and genomic information. Animals were genotyped with either the 50K SNP panel or the low-density panel. Imputation to 50K SNP for animals genotyped with the low-density panel was done using FImpute software. Digital dermatitis was analyzed as a binary trait, where 1 was assigned to the presence of the lesion. Traditional pedigree-based and genomic evaluations were computed using the BLUPF90 software. The daughter deviation (DD2016) from complete data set was used as the expected future progeny performance of validation bulls to assess predictive ability based on the reduced data set, in which only phenotypes recorded up to 2011 were included. The coefficient of determination (R2) from the regression of DD2016 on parent average or GEBV in 2011 was used to assess the validation reliability of the models. Digital dermatitis had a prevalence of 18.5% and a heritability (SE) of 0.08 (0.005). As expected, predictions based on genomic information had higher R2 than traditional evaluations (0.11 and 0.07, respectively, for sires with more than 10 daughters with recorded lesions in the complete data set). Also, the inflation of the evaluation for bulls in the validation population was lower when genomic information was included in the analyses. For sires with more than 10 daughters with recorded lesions in the complete data set, the regression coefficients for traditional and genomic evaluation were 0.86 and 0.99, respectively. Results indicate that the use of genomic selection for digital dermatitis can improve the accuracy of selection for young bulls compared with using selection based on parent average information.
Key Words: digital dermatitis, single-step
