Comparative genomics allows the transfer of genomic information from a well-characterized species to another that is less well described. It can be applied at all levels from that of the chromosome to the genome sequence. However, despite the recent advances in sequencing technologies, the considerable effort involved in producing a genome sequence assembly is reflected by the small number of vertebrate genomes that have been sequenced to date. In birds, there is only one published genome sequence, that of the chicken [1
], with the zebra finch genome due to be published soon.
Combining cross-species fluorescent in-situ hybridization (FISH) and microarray analysis using resources developed in the chicken provides a powerful tool for the identification of gross genomic rearrangements, gene gains/losses, copy number variants (CNVs) and gene order in other bird species. These techniques do not require sequence data for any species other than the reference (i.e. chicken). We have previously successfully used this approach for a genome wide comparison of chromosomal rearrangements and CNVs between chicken and turkey[2
]. This revealed a strong conservation of genome structure over about 30 million years of evolution between chicken and turkey[3
]. In particular, our results suggested that, when compared to mammalian genomes, bird genomes contain a low number of CNVs (i.e. polymorphisms in the number of copies of a DNA fragment 1 kb or larger[4
], with the exception of insertions or deletions of transposable elements[5
]). The latter finding indicates that patterns of CNVs in bird genomes mirror the low number of chromosomal rearrangements in this phylogenetic group[2
Following on from the turkey, the Pekin duck (Anas platyrhynchos
, APL) is the next obvious target among domestic birds for detailed genomic studies due to its agricultural importance, with worldwide duck consumption being between 4 and 5% of the total poultry market[7
]. Duck is also an important target for immunological studies because of its resistance to avian influenza[8
]. Despite this, genomic information about the duck is limited to a few linkage and physical mapping studies. Huang et al
] produced a preliminary genetic map based on 240 microsatellite loci and assigned 11 out of 19 linkage groups to ten duck (APL) chromosomes by FISH mapping of 28 BACs. Cross-species chromosome painting and G-banding studies [10
] have suggested one interchromosomal difference between the chicken and duck karyotypes – the ancestral chromosomes 4 and 10, fused in the chicken lineage to give GGA4q and GGA4p respectively, remain separate in duck[6
]. This interchromosomal rearrangement presumably explains the difference in diploid chromosome number between the two species, which is 2n = 78 in chicken and 2n = 80 in duck. FISH mapping of 57 chicken BACs revealed small intrachromosomal rearrangements in APL2, 7, 8 and Z and confirmed synteny for GGA9, 11, 13–15, 18 and 28 in the duck genome[13
]. However, no molecular markers are available for the remaining microchromosomes, which are indistinguishable by conventional cytogenetics. It is also unclear which duck chromosome corresponds to GGA4p (ancestral chromosome 10). Thus, from a molecular cytogenetic standpoint, the duck genome is at present only partially defined, and given the low number of physical markers mapped by FISH, it is possible that hitherto undetected intrachromosomal rearrangements exist.
No data are currently available concerning CNVs in duck or indeed any other bird species than chicken and turkey. CNVs have been found to contribute significantly to normal and disease-related genetic and phenotypic variation in humans and other primates[5
]. Studies of the evolutionary significance of CNVs have largely focused on primates and revealed numerous lineage-specific gene gains and losses and CNVs (e.g. [15
Our previous study of CNVs in chicken and turkey revealed a total of 16 CNVs[2
]. Five of these CNVs appear to be shared in layer and broiler chickens, and in turkey, at regions dubbed "CNV hotspots" (i.e. genomic regions in which CNVs of approximately equal size were found in both chicken breeds and in turkey). Given that the contribution of CNVs to phenotypic variation is becoming increasingly clear, analyses of this kind of structural variation in organismic groups other than mammals are clearly needed.
Here, we present a detailed molecular cytogenetic map for the duck based on comparative FISH mapping of 155 chicken BACs, which revealed several hitherto undescribed intrachromosomal rearrangements. We also provide an analysis of CNVs in the duck genome by array comparative genomic hybridisation (array CGH) of duck DNA to a commercially available chicken whole-genome oligonucleotide tiling path microarray. The analysis of CNVs supports the hypotheses that bird genomes contain fewer CNVs than mammalian genomes and that some CNVs appear to be consistently shared across species, forming CNV hotspots.