Journal
JOURNAL OF ANIMAL SCIENCE
Volume 87, Issue 2, Pages 507-515Publisher
AMER SOC ANIMAL SCIENCE
DOI: 10.2527/jas.2008-1092
Keywords
connectedness; genetic evaluation; prediction error; sheep
Categories
Ask authors/readers for more resources
Connectedness among animals in separate flocks reduces the risk of biased comparisons when selecting across flocks on EBV. The objective in this study was to assess levels of connectedness in the genetic evaluation of weaning weight among Targhee and Suffolk flocks participating in the US National Sheep Improvement Program (NSIP). Among flocks currently participating in the NSIP, a total of 25,404 weaning weight and 35,794 pedigree records were available for 16 Targhee flocks, and 14,017 weaning weight and 18,311 pedigree records were available for 24 Suffolk flocks. Connectedness was measured by using 2 different methods. First, numbers of progeny with recorded weaning weights from linking sires (defined as sires with progeny in multiple flocks or sires born in one flock with progeny in another flock) were counted. Second, connectedness was measured by calculating the average prediction error correlation of mean flock EBV (flock r(ij)). Benchmarks for flock r(ij) were established, with 0.10 and 0.05 representing low and moderate risk of bias associated with comparing EBV among flocks, respectively. From 1995 through 2004, 44% of Targhee lambs with weaning weights were born to linking sires; in Suffolk lambs, that value was 23%. In 1990, 1995, and 2005, average flock r(ij) were 0.10, 0.19, and 0.28, respectively, among Targhee flocks, and 0.02, 0.02, and 0.04, respectively, among Suffolk flocks that participated in NSIP in all 3 yr. Among all active flocks in 2005, flock r(ij) averaged 0.13 in Targhees and 0.03 in Suffolks. Hierarchical clustering of flocks based on flock r(ij) revealed that all active Targhee flocks connected at a level near or above 0.10. In Suffolk flocks, 2 distinct clusters had formed, in which connectedness was relatively high within each cluster (flock r(ij) near 0.10) but was near zero between clusters. Risk of bias in comparing EBV among flocks in the Targhee was low; however, caution should be exercised when comparing EBV between Suffolk flocks from different clusters.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available