An ABC estimate of pedigree error rate: application in dog, sheep and cattle breeds.

On the basis of correlations between pairwise individual genealogical kinship coefficients and allele sharing distances computed from genotyping data, we propose an approximate Bayesian computation (ABC) approach to assess pedigree file reliability through gene-dropping simulations. We explore the features of the method using simulated data sets and show precision increases with the number of markers. An application is further made with five dog breeds, four sheep breeds and one cattle breed raised in France and displaying various characteristics and population sizes, using microsatellite or SNP markers. Depending on the breeds, pedigree error estimations range between 1% and 9% in dog breeds, 1% and 10% in sheep breeds and 4% in cattle breeds.

Genetic variability, structure and assignment of Spanish and French pig populations based on a large sampling.

The Spanish and French pig populations share the common practice of quasi systematic paternity control of pure breed and composite line males. Ten microsatellite markers are in common between Spain and France controls, among the 17 markers used in France and the 13 used in Spain. After the adjustment of allele sizes, it is possible to merge the two datasets and to obtain a set of 5791 animals, including the vast majority of the males in the Duroc, Landrace, Large White and Piétrain French and Spanish breeds. Twelve French composite lines are also available. The genetic diversity analysis of these pig populations is presented, as well as the assignment of an individual to its breed. The effects of heterogeneous sampling across time and of relatedness among animals are also assessed. Consistent with the results of the previous studies, we found that different populations from the same breed clearly clustered together. In addition, all populations of this study, whether purebred or composite, are quite well differentiated from the other ones. As a result, we note that the 10 microsatellites commonly used for paternity control ensure a powerful detection of the breed of origin, with the power of detection being 95-99%. The detection of the exact population within breed is more difficult, but the power exceeds 70% for most of the populations. Practical implications include, for instance, the detection of outlier animals, crosses and admixture events.

Genetic diversity of dog breeds: between-breed diversity, breed assignation and conservation approaches.

Genetic relationships between 61 dog breeds were investigated, using a sampling of 1514 animals and a panel of 21 microsatellite markers. Based on the results from distance-based and Bayesian methods, breed constituted the main genetic structure, while groups including genetically close breeds showed a very weak structure. Depending on the method used, between 85.7% and 98.3% of dogs could be assigned to their breed, with large variations according to the breed. However, breed heterozygosity influenced assignment results differently according to the method used. Within-breed and between-breed diversity variations when breeds were removed were highly negatively correlated (r = -0.963, P < 0.0001), because of the genetic structure of the breed set.

Genetic diversity of dog breeds: within-breed diversity comparing genealogical and molecular data.

The genetic diversity of 61 dog breeds raised in France was investigated. Genealogical analyses were performed on the pedigree file of the French kennel club. A total of 1514 dogs were also genotyped using 21 microsatellite markers. For animals born from 2001 to 2005, the average coefficient of inbreeding ranged from 0.2% to 8.8% and the effective number of ancestors ranged from 9 to 209, according to the breed. The mean value of heterozygosity was 0.62 over all breeds (range 0.37-0.77). At the breed level, few correlations were found between genealogical and molecular parameters. Kinship coefficients and individual similarity estimators were, however, significantly correlated, with the best mean correlation being found for the Lynch & Ritland estimator (r = 0.43). According to both approaches, it was concluded that special efforts should be made to maintain diversity for three breeds, namely the Berger des Pyrénées, Braque Saint-Germain and Bull Terrier.

An international parentage and identification panel for the domestic cat (Felis catus).

Seventeen commercial and research laboratories participated in two comparison tests under the auspices of the International Society for Animal Genetics to develop an internationally tested, microsatellite-based parentage and identification panel for the domestic cat (Felis catus). Genetic marker selection was based on the polymorphism information content and allele ranges from seven random-bred populations (n = 261) from the USA, Europe and Brazil and eight breeds (n = 200) from the USA. Nineteen microsatellite markers were included in the comparison test and genotyped across the samples. Based on robustness and efficiency, nine autosomal microsatellite markers were ultimately selected as a single multiplex 'core' panel for cat identification and parentage testing. Most markers contained dinucleotide repeats. In addition to the autosomal markers, the panel included two gender-specific markers, amelogenin and zinc-finger XY, which produced genotypes for both the X and Y chromosomes. This international cat parentage and identification panel has a power of exclusion comparable to panels used in other species, ranging from 90.08% to 99.79% across breeds and 99.47% to 99.87% in random-bred cat populations.

Genetic diversity within and among four South European native horse breeds based on microsatellite DNA analysis: implications for conservation.

In the present study, genetic analyses of diversity and differentiation were performed on four Basque-Navarrese semiferal native horse breeds. In total, 417 animals were genotyped for 12 microsatellite markers. Mean heterozygosity was higher than in other horse breeds, surely as a consequence of management. Although the population size of some of these breeds has declined appreciably in the past century, no genetic bottleneck was detected in any of the breeds, possibly because it was not narrow enough to be detectable. In the phylogenetic tree, the Jaca Navarra breed was very similar to the Pottoka, but appeared to stand in an intermediate position between this and the meat breeds. Assuming that Pottoka is the breed less affected by admixture, the others gradually distanced themselves from it through varying influences from outside breeds, among other factors. In a comparative study with other breeds, the French breeds Ardanais, Comtois, and Breton were the closest to the four native breeds. Three different approaches for evaluating the distribution of genetic diversity were applied. The high intrabreed variability of Euskal Herriko Mendiko Zaldia (EHMZ) was pointed out in these analyses. In our opinion, cultural, economic, and scientific factors should also be considered in the management of these horse breeds.

Which insemination results in fertilization when several are performed before ovulation?

The aim of the present study was to determine which artificial insemination results in fertilization when mares are inseminated several times before ovulation. Mares in oestrus were inseminated over 62 cycles with fresh semen at 48 h intervals from when a follicle > or =30 mm in diameter was detected until ovulation. The number of inseminations was limited to three. Three fertile stallions were used and a different stallion was used for each artificial insemination. The order of the three stallions was changed for each cycle. Embryos were collected between day 10 and day 12 after ovulation and paternity was checked using DNA typing. When two inseminations were performed per cycle, 14 of 17 embryos were the result of the insemination performed on days 2-4 before ovulation and three embryos were the result of the insemination performed on days 0-2. When three inseminations were performed, 1 of 6, 2 of 6 and 3 of 6 embryos resulted from the inseminations performed 4-6, 2-4 and 0-2 days before ovulation, respectively. Thus, 17 of 23 (74%) oocytes were not fertilized as a result of the insemination performed 0-2 days before ovulation. The mean interval between fertilization and ovulation in the mares from which embryos were recovered and tested (n=23) was 2.6 +/- 1.0 days. These results indicate that spermatozoa can remain viable in the genital tract of mares for at least 2.6 days.

Report of the International Equine Gene Mapping Workshop: male linkage map.

The goal of the First International Equine Gene Mapping Workshop, held in 1995, was the construction of a low density, male linkage map for the horse. For this purpose, the International Horse Reference Family Panel (IHRFP) was established, consisting of 12 paternal half-sib families with 448 half-sib offspring provided by 10 laboratories. Blood samples were collected and DNA extracted in each laboratory and sent to the Lexington laboratory (KY, USA) for dispatch in aliquots to 14 typing laboratories. In total, 161 markers (144 microsatellites, seven blood groups and 10 proteins) were tested for all families for which the sire was heterozygous. Genealogies and typing data were sent for analysis to the INRA laboratory (Jouy-en-Josas, France) according to a specific format and entered into a database with input verification and output processes. Linkage analysis was performed with the CRIMAP program. Significant linkage was detected for 124 loci, of which 95 were unambiguously ordered using a multipoint analysis with an average spacing of 14.2 CM. These loci were distributed among 29 linkage groups. A more comprehensive analysis including synteny group data and FISH data suggested that 26 autosomes out of 31 are covered. The complete map spans 936 CM.

Characterization, genetic and physical mapping analysis of 36 horse plasmid and cosmid-derived microsatellites.

Thirty-six new horse microsatellites (11 from plasmid libraries and 25 from a cosmid library) were isolated and characterized on a panel of four horse breeds. Thirty were found to be polymorphic with heterozygosity levels ranging between 0.20 and 0.87. Twenty-two of the cosmids were physically mapped to R-banded single horse Chromosomes (Chrs) 1, 3, 4, 9, 11, 12, 13, 15, 18, 19, 21, 22, 23 and three to pericentromeric regions. Furthermore, linkage analysis between a selection of 42 DNA markers, including those presented in this study, and 16 conventional markers of the horse hemotype was performed on six paternal half-sib horse families. Five linkage groups were detected, of which four were assigned to Chr 10, 11, 15, and 18. This work increased by one-third the number of published polymorphic DNA markers suitable for horse mapping and approximately doubled the number of known linkage groups. Our cosmids labeled 14 out of the 31 horse autosomes. Moreover, the physical anchoring of part of these markers will orient linkage and synteny groups on the chromosomes and will contribute to their assignment.

Genetic blood markers in Arabian, Barb and Arab-Barb horses in Morocco.

Gene frequencies at 16 blood group and protein polymorphism loci (A, C, D, K, P, Q, U, Al, Gc, Es, A1B, Tf, PGD, PGM, GPI and Pi) are given for three horse breeds in Morocco (Arabian, Arab-Barb and Barb). These data are used to calculate average heterozygosity (h), Nei's standard genetic distance (DN) and probability of exclusion (PE). Variability expressed as the average heterozygosity was lower in the Arabian (0.330 +/- 0.066), while it was higher and almost the same in the Arab-Barb (0.413 +/- 0.071) and the Barb (0.414 +/- 0.070). The shortest genetic distance was found between Barb and Arab-Barb. The 16 loci used are at least 95% effective for recognizing incorrect paternity in these breeds. The Barb and Arab-Barb genetic profiles obtained showed the rare variants interesting perhaps in the context of European and American breeds: notably Dcfgkm, Ddekl, Es-N, Tf-A and Pi-W.

[Monoclonal antibodies directed against equine blood group antigens]. / Anticorps monoclonaux dirigés contre les antigènes des groupes sanguins des chevaux.

The chief application of blood typing in domestic animals is in the verification of parentage. However, the acquisition of good standardized reagents in sufficient quantity remains an obstacle for the development of this work. The production of monoclonal antibodies directed against blood group determinants offers an attractive means of improving both the quality and quantity of serological reagents, and could facilitate the definition of new specificities. Fusions between a mouse myeloma line and splenocytes from mice immunized with horse red cells have resulted in four hybridomas producing antibodies against equine erythrocyte groups. Two are directed against the established groups Aa and Ca, while one reacts with a sub-group of De, and another, still under study, appears to be anti-Di. The anti-Aa and anti-Ca monoclonals have high affinity and fix complement, and are now in routine use as blood grouping reagents. This is remarkable since good antibodies to these specific sites are virtually impossible to obtain by allo-immunization. These results offer encouragement for the future production of monoclonal antibodies against red cell and lymphocyte antigens in the domestic species.

A contribution to the D system in horses.

The inheritance of a new D system red cell antigen, factor 22, is described. It has also been possible to discriminate more efficiently between D system phenogroups enabling genotypes to be identified from phenotypes in the majority of cases. This improves the accuracy of animal identification and gene frequency estimates.