Each species offers a different lens through which to examine the evolutionary past. The puffer fish, studied by a team representing, among other institutions, Genoscope in France and the Broad Institute at MIT and Harvard, represents a lineage that split from our own perhaps 450 million years ago. A close examination of its sequence reveals two interesting discoveries. First, at some point in the history of the ray-finned fishes the genome was duplicated in its entirety. This had been inferred from previous studies of particular genes; the draft sequence confirms it. One of the confirmations involved comparing the positions of about 6,500 genes in puffer fish and human. In almost every case, a stretch of genes in human could be located at two positions in puffer fish. Further analysis of the genes in puffer fish and human allowed the team to deduce that our common ancestor, which has been extinct for 450 million years or thereabouts, had 12 pairs of chromosomes and to recreate the structure of those chromosomes in some detail.

Jump forward about 140 million years and you get to the last common ancestor of human and chicken, sequenced by the International Chicken Genome Sequencing Consortium (ICGSC). The chicken genome sheds light on one of the least understood aspects of the human genome: non-coding elements. Sequences that are conserved across lineages may indicate functionally important genes, but if the lines have not been separated for long the similarities may simply reflect a lack of time for changes to have occurred. The evolutionary distance between chicken and human strongly suggests that similarities do represent functionally important areas. The ICGSC identified 70 million base pairs that are highly conserved between chicken and human. Almost a third of these lie between known genes, in non-coding regions, most of them rather distant from the nearest genes. Some of these represent known regulators of gene function, while others are probably new kinds of regulators that have yet to be identified.

And then there is the chimpanzee, split from ourselves maybe 6 million years ago and analyzed by the Chimpanzee Sequencing and Analysis Consortium (CSAC). As the CSAC points out, the short time since our lines diverged means that almost all of the sequence is identical by descent. Rather than focussing on similarities, then, like the more distant comparisons such as chicken and puffer fish, the focus is on differences, which presumably underlie some of the differences between human and chimp. The old and now somewhat hoary figure of "99% similar" holds up, with 98.7% of the sequence identical. Do the differences help us to understand what makes us human?

One way to approach the question is to look in genes at the ratio between synonymous substitutions, which do not change the structure of the protein the gene codes for, and non-synonymous substitutions, which do affect the protein. Regions subject to strong constraining selection will have an excess of synonymous over non-synonymous substitutions, while weak selection, or indeed positive selection, will favor non-synonymous substitutions. Genes related to brain function show no evidence of positive selection in humans, and the most highly selected genes are those related to the immune system and to reproduction. Protein evolution itself is probably not that important in the evolution of human differences.

Loss of function could also contribute to the differences. Humans have less body hair, we mature more slowly, and our skulls keep expanding for longer. The CSAC identified 53 human genes that have disruptive insertions or deletions compared to the chimpanzee. Some are already hinting at important changes that might underlie our "humanness."