What Makes us Different?

Still, the principle of gene-by-gene comparison remains a powerful one, and just a year ago geneticists got hold of a long-awaited tool for making those comparisons in bulk. Although the news was largely overshadowed by the impact of Hurricane Katrina, which hit the same week, the publication of a rough draft of the chimp genome in the journal Nature immediately told scientists several important things. First, they learned that overall, the sequences of base pairs that make up both species' genomes differ by 1.23%--a ringing confirmation of the 1970s estimates--and that the most striking divergence between them occurs, intriguingly, in the Y chromosome, present only in males. And when they compared the two species' proteins--the large molecules that cells construct according to blueprints embedded in the genes--they found that 29% of the proteins were identical (most of the proteins that aren't the same differ, on average, by only two amino-acid substitutions).

The genetic differences between chimps and humans, therefore, must be relatively subtle. And they can't all be due simply to a slightly different mix of genes. Even before the human genome was sequenced back in 2000, says biologist Sean Carroll of the University of Wisconsin, Madison, "it was estimated that humans had 100,000 genes. When we got the genome, the estimate dropped to 25,000. Now we know the overall number is about 22,000, and it might even come down to 19,000."

This shockingly small number made it clear to scientists that genes alone don't dictate the differences between species; the changes, they now know, also depend on molecular switches that tell genes when and where to turn on and off. "Take the genes involved in creating the hand, the penis and the vertebrae," says Lovejoy. "These share some of the same structural genes. The pelvis is another example. Humans have a radically different pelvis from that of apes. It's like having the blueprints for two different brick houses. The bricks are the same, but the results are very different."

Those molecular switches lie in the noncoding regions of the genome--once known dismissively as junk DNA but lately rechristened the dark matter of the genome. Much of the genome's dark matter is, in fact, junk--the residue of evolutionary events long forgotten and no longer relevant. But a subset of the dark matter known as functional noncoding DNA, comprising some 3% to 4% of the genome and mostly embedded within and around the genes, is crucial. "Coding regions are much easier for us to study," says Carroll, whose new book, The Making of the Fittest: DNA and the Ultimate Forensic Record of Evolution, delves deep into the issue. "But it may be the dark matter that governs a lot of what we actually see."