John Hawks talks about some interesting complexities coming out of human genome studies. In a series of posts (for example here and here ) he discusses data suggesting that Homo Sapiens did not originate as a single interbreeding tribe (they were not panmictic ). Rather, there are portions of our genomes-- plural intended!-- which have been genetically separated for even a million years, i.e. last recombined before Homo Sapiens was a species.
The evidence is coming from the avalanche of high resolution data about human genomic variation. Over time, genome differences can occur in any organism via mutations, and if they do not interfere with survival, they can be kept in the descendants. In most cases, genetic mutations which occur far enough apart on a chromosome will tend to get exchanged between pieces of DNA, via a process called recombination . What is critical here is that recombination only occurs between chromosomes inside the same individual. If a group with a new variation never interbreeds with the main group, the variant and original chromosome will never "see" each other, and recombination will not occur.
Several groups studying small chromosomal regions (about 20 kilobases of sequence) find that tandem strings of differences have been preserved, thus that recombination has not occured. In the second case discussed by Hawks, two individuals in Africa were found to have a ten-step difference in a region of the X-chromosome relative to the more common allele. Recombination among the 10 differences had not occurred, even though the more common allele was obviously capable of recombination with other versions.
So, as Hawks says, the very low likelihood of observing this result under panmixia indicates that it is likely that parts of the ancestral population were out of genetic contact for some period of time. Additional data would be required to place these observations along the continuum from a highly dispersed population to admixture (breeding) between fully differentiated subspecies. The authors of the study are going for the latter end, and Hawks thinks support for this idea may reach a tipping point in 2005. The authors say "This inference supports human evolution models that incorporate admixture between divergent African branches of the genus Homo--" i.e., it is consistent with the idea that genetically divergent hominids interbred within the ancestry of Homo sapiens. In any case, the clean Out-of-Africa picture of a single interbreeding tribal origin of Homo Sapiens is in trouble.