An important prerequisite for a conservation programme is a comprehensive description of genetic diversity. The aim of this study was to use anonymous genetic markers to assess the between- and the within-population components of genetic diversity for European pig breeds at the scale of the whole continent using microsatellites. Fifty-eight European pig breeds and lines were analysed including local breeds, national varieties of international breeds and commercial lines. A sample of the Chinese Meishan breed was also included. Eleven additional breeds from a previous project were added for some analyses. Approximately 50 individuals per breed were genotyped for a maximum of 50 microsatellite loci. Substantial within-breed variability was observed, with the average expected heterozygosity and observed number of alleles per locus being 0.56 [range 0.43-0.68] and 4.5 respectively. Genotypic frequencies departed from Hardy-Weinberg expectations (P < 0.01) in 15 European populations, with an excess of homozygotes in 12 of them. The European breeds were on average genetically very distinct, with a Wright F(ST) index value of 0.21. The Neighbour-Joining tree drawn from the Reynolds distances among the breeds showed that the national varieties of major breeds and the commercial lines were mostly clustered around their breeds of reference (Duroc, Hampshire, Landrace, Large White and Piétrain). In contrast, local breeds, with the exception of the Iberian breeds, exhibited a star-like topology. The results are discussed in the light of various forces, which may have driven the recent evolution of European pig breeds. This study has consequences for the interpretation of biodiversity results and will be of importance for future conservation programmes.

Despite the chicken being one of the most genetically mapped of all animals, its karyotype remains poorly defined. This is primarily due to microchromosomes that belie assignment by conventional methods. To address this problem, we have developed chromosome-specific paints using flow cytometry and microdissection. For the microchromosomes it was necessary to amplify and label DNA from single microdissected chromosomes.

Reciprocal chromosome painting between mouse and rat using complete chromosome probe sets of both species permitted us to assign the chromosomal homology between these rodents. The comparative gene mapping data and chromosome painting have a better than 90% correspondence. The reciprocal painting results graphically show that mouse and rat have strikingly different karyotypes. At least 14 translocations have occurred in the 10–20 million years of evolution that separates these two species. The evolutionary rate of chromosome translocations between these two rodents appears to be up to 10 times greater than that found between humans and cats, or between humans and chimpanzees, where over the last 5–6 million years just one translocation has occurred. Outgroup comparison shows that the mouse genome has incorporated at least three times the amount of interchromosomal rearrangements compared to the rat genome. The utility of chromosome painting was also illustrated by the assignment of two new chromosome homologies between rat and mouse unsuspected by gene mapping: between mouse 11 and rat 20 and between mouse 17 and rat 6. We conclude that reciprocal chromosome painting is a powerful method, which can be used with confidence to chart the genome and predict the chromosome location of genes. Reciprocal painting combined with gene mapping data will allow the construction of large-scale comparative chromosome maps between placental mammals and perhaps other animals.

Childbirth is a period of substantial rapid biological and psychological change and a wide range of psychotic disorders can occur ranging from mild 'baby blues' to severe episodes of psychotic illnesses. Puerperal psychosis is the most extreme form of postnatal psychosis, occurring in 1 in 1,000 births. In this study, we have used the pig as an animal model for human postnatal psychiatric illness. Our aim was to identify quantitative trait loci (QTL) associated with maternal (infanticide) sow aggression. This is defined by sows attacking and killing their own newborn offspring, within 24 hr of birth. An affected sib pair whole genome linkage analysis was carried out with 80 microsatellite markers covering the 18 porcine autosomes and the X chromosome, with the aim of identifying chromosomal regions responsible for this abnormal behavior. Analysis was carried out using the non-parametric linkage test of Whittemore and Halpern, as implemented in the Merlin software. The results identified 4 QTL mapping on Sus scrofa chromosomes 2 (SSC2), 10 (SSC10), and X (SSCX). The peak regions of these QTL are syntenic to HSA 5q14.3-15, 1q32, Xpter-Xp2.1, and Xq2.4-Xqter, respectively. Several potential candidate genes lie in these regions in addition to relevant abnormal behavioral QTL, found in humans and rodents.

The new data reported in this study reveal further discrete X chromosome intervals not previously associated with the disease and therefore implicate new clusters of candidate genes. Further studies will be required to elucidate their involvement in POF.