The genetic diversity and structure of 18 sheep breeds exposed to isolation and selection.

The phylogenetic layout of the genotyped (30 microsatellite) 18 sheep breeds in this study demands and provides the opportunity to evaluate both neutral and adaptive components of genetic diversity in a naturally and artificially selected and subdivided sheep population. Seven Pramenka strains from Bosnia and Herzegovina and Croatia characterized by a very low intensity of artificial selection, preserved the highest neutral genetic variability. Eight central and north-western European breeds under considerable artificial isolation and selection preserved the lowest genetic variability. Only combinations of various phylogenetic parameters offer a reasonable explanation for underlying evolutionary forces working in the investigated island and mainland sheep breeds under variable natural and artificial selection. More than 60% of total genetic, diversity was allocated to virtually unselected Pramenka strains, and an additional 25% to native moderately selected Graue Gehoernte Heidschnucke and intensively selected Ostfriesische Milchschafe. Some economically very important breeds and strains did not contribute to a pool with maximal genetic diversity, while they play an important role in the cultural heritage of respective countries.

Genetic diversity and structure of the West Balkan Pramenka sheep types as revealed by microsatellite and mitochondrial DNA analysis.

Several different phenotypes of the native Pramenka sheep have been developed in the Balkan region for different environmental and socio-cultural conditions. Animals from seven West Balkan Pramenka sheep types were analysed for 15 microsatellite markers and for mitochondrial DNA (mtDNA) and the results were used to assess genetic variation within and among the types and to infer the genetic population structure of the Pramenka sheep. Mean expected heterozygosity and allelic richness over the microsatellite loci and sheep types were 0.78 and 7.9, respectively. A Bayesian statistical method for estimating hidden genetic structure suggested that a core of the largest panmictic population was formed by Serbian, Kosovan, Bosnian, Montenegrin and Albanian types, while Croatian and Macedonian types comprised two other main populations, respectively. Mitochondrial DNA analysis revealed two mtDNA haplogroups in the Pramenka sheep, B and A, with a frequency of 93.7% and 6.3%, respectively. A total of 60 mtDNA haplotypes were found in 64 animals sequenced, and the mean nucleotide and haplotypic diversities over the types were 0.013 and 0.945, respectively. Molecular analysis suggests that the West Balkan Pramenka sheep types have their origins in two distinct maternal lineages of domestic sheep and different Pramenka phenotypes tend to form few panmictic populations. The Pramenka sheep represents a valuable resource of genetic diversity in sheep.

Frequency and heritability of supernumerary teats in German simmental and German Brown Swiss cows.

The incidence of supernumerary teats has been recorded in 179,793 German Simmental and 37,460 German Brown Swiss cows. Data were collected from first-crop daughters of test bulls from 1987 to 1998. The number of sires was 4,298 and 1,039, respectively. The average frequency of affected animals was 44.3% in German Simmental and 31.2% in German Brown Swiss. A significant yet small effect was found for herd-book membership (yes/no) and, in the Simmental data only, for inspector. The impact of year of birth and year of inspection was also significant, and more important, reflected a decrease of the population averages with time. Surprisingly, the incidence of supernumerary teats increased significantly with the parity number of dams in both breeds. A Bayesian threshold animal model approach was used to estimate the heritability of the occurrence of supernumerary teats. The posterior mean for the heritability was h2 = 0.45 in German Simmental and h2 = 0.43 in Brown Swiss, with standard errors of 0.01 and 0.03, respectively. Ranks of sires obtained from threshold and linear models showed a rank correlation of roughly 0.8 in both breeds. For a sound identification of the worst sires, a threshold model is recommended.