The chicken has a haploid genome size of around 1.2 × 10⁹ base-pairs, around 40% the size of mammals; it is estimated that the genome sequence contains around 20,000-23,000 genes 3, a slightly smaller number than in mammals 567. Many of the genes have been mapped to chromosomes, and these maps can be compared to other genomes to discover syntenic regions, where the same genes occur in a similar order along the chromosomes of different organisms. This does not just allow analysis of gene order in the chicken itself; the chicken genome can be used as an outgroup to the human and mouse genomes, allowing rates of gene rearrangement in the human genome and the architecture of the ancestral mammalian genome to be investigated. This approach uncovers a number of interesting features 3. The rate of rearrangement in the human lineage is very slow compared to that of mouse, and that inferred for the mammalian common ancestor is slower still. When a fish out-group is added, the analysis reveals that the rate of rearrangement on the chicken lineage is comparable to that of the mammalian common ancestor 3. This supports a previous observation that synteny is more conserved between human and chicken than it is between human and mouse 8, and suggests that the stability of the chicken genome makes it a good candidate for future studies of vertebrate genome architecture.

The chicken gene set can also be compared with those of mammalian genomes to discover lineage-specific changes to protein-coding genes or gene families, such as duplication or loss. In many cases, these changes mirror phenotypic change. For example, mammals appear to have lost several genes associated with egg production, in particular the avidin gene family 3. These genes encode egg-white proteins and have homologs in invertebrates, indicating that they have been lost in mammals, probably in association with the reduction in egg size and internalization of the embryo on this lineage. The chicken genome appears to have fewer innovations and an enhanced rate of loss compared with other animal genomes 3. Because the genome sequence is not finished and no other diapsid genomes are available for comparison, specific losses on this lineage cannot be discussed with confidence, but gain (or duplication) of genes can be determined with more certainty.

Gene-family expansion plays a substantial role in lineage-specific evolution. For example, both mammals and chickens have expanded their keratin gene repertoire by gene duplication, but in quite different directions 3. Birds use a large, avian-specific family of keratin genes to form proteins for scales, claws and feathers. Mammals have undergone an independent expansion of a different keratin family, which is used to form hair fibers. A more surprising finding is that chickens have at least 218 non-identical genes that are orthologous to the human OR5U1 and OR5BF1 olfactory receptor genes 3. Not only is this an exceptionally large expansion, but it is traditionally thought that birds have a poor sense of smell 9! The chicken genome sequence reveals that, thanks to this expansion, chickens have a similar number of olfactory receptor genes to humans 510, suggesting that their sense of smell may play more of a part in their behavior than previously thought.

Comparisons with human and pufferfish (Takifugu rubripes) reveal around 7,000 chicken genes that have 1:1 orthologs in both species, suggesting a 'core' of genes that may have an essential role in all vertebrates 3. The sequences in this core tend to be more conserved than other human/chicken orthologs, indicating that strong purifying selection is acting upon them, furthering the case for their functional importance. The results also suggest that these are genes that are expressed in many different tissues; this is not unexpected, as previous mammalian studies have suggested that rapidly evolving genes are expressed in fewer tissues 1112. The chicken genome 3 supports this theory: genes that can be found as expressed sequence tags (ESTs) from many tissues tend to be well conserved between human and chicken, whereas those expressed in few tissues are more divergent. The authors 3 also imply that a high proportion of the core genes are involved in cytoplasmic and nuclear functions, such as protein and intracellular transport. It would be interesting to discover how many of these core genes have previously been defined as mammalian housekeeping genes 13. It should also be possible to examine whether any of these genes are also conserved across the invertebrates, to determine whether there is an animal-specific core of genes and how this differs from the vertebrate-specific core. It is generally accepted that an enhanced repertoire of developmental genes has played a role in the many innovations on the vertebrate lineage 14, but comparisons of this housekeeping dataset with invertebrate genomes - such as those of the fruit fly or sea squirt - could provide evidence for other sources of vertebrate novelty.

The chicken genome sequence assembly is currently estimated to cover 97% of the genome 3. It is still very much in the draft phase, and a great deal of future work is likely to be necessary to refine the data. Despite the incompleteness of the protein-coding dataset, many new observations can be made about the structure and content of the avian gene set and how it compares with mammalian genomes. Analysis of the chicken genome also highlights the importance of sequencing genomes that lie in key positions on the tree of life: complete sequences of genomes from across the vertebrates, for example, would allow us to reconstruct the genome architectures of species at each node along this lineage. Closely related genomes can also reveal much, as in the case of rat and mouse genome analyses 715. But no matter what organism it comes from, each new genome sequence has a fascinating story to tell, and adds more detail to our knowledge of genome evolution and organization.