September 16, 2012

Longer time scale for human evolution (Hawks 2012)

Scally and Durbin published a recent review on the implications of a slower human autosomal mutation rate, and now John Hawks has a commentary on the same topic in PNAS (pdf; paywall). He goes through a lot of the evidence of early fossil hominins and ape and mentions several examples that harmonize with the slower mutation rate. As expected, he also finds a better agreement of the slow mutation rate with the evidence for Neandertals where 530,000 year old finds from Atapuerca show signs of belonging to the Neandertal lineage, a date that is inconsistent with a late divergence of modern humans and Neandertals. Finally, he has this to say about modern humans:

Across this same time scale, the archaic ancestors of today’s Africans had already developed an intricate population structure. Genomic investigation of African hunter–gatherers has opened new windows onto this deep genetic history of differentiation and introgression (14, 15), bringing the origin of modern African diversity into the population structure of the early Middle Pleistocene. A simple hypothesis of modern human origins in a bottlenecked population cannot account for this diverse genetic history.    
The mtDNA time scale now poses a hanging question. Mitochondrial mutations occur much more often than nuclear DNA mutations, with greater heterogeneity among sites (16). Still, our estimate of mtDNA substitution rates depends on our estimates of branch lengths of the primate phylogeny. Until now, mitochondrial comparisons have been the strongest evidence in favor of a short time scale for the dispersal and differentiation of non-African peoples, within the past 70,000 y (17). Some recent attempts to examine the relationships of non-African populations using nuclear genome data have led to time scales in excess of 100,000 y (18), and others favor more recent estimates (19). Despite the recency of this work, most authors have continued to use an outdated fast molecular clock and short generation time estimates. As we move forward, such results will need to be corrected or adjusted to enable comparisons with current work. 
There is a very interesting question here, which I've mentioned before, but is worth repeating: admixture between divergent lineages can inflate split times. Acceptance of the slow autosomal mutation rate will result in split times in excess of 100 thousand years for Africans vs. non-Africans, and perhaps 300 thousand years for African hunter-gatherers. On the other hand, the mtDNA clock (haplogroup L3 = 70ky), no matter how it is recalibrated is unlikely to match these old dates, and the Y-chromosome clock (current estimate of its root a little more than 100 ky, and of the dominant African lineage E on the cusp of the LSA) will certainly not match them.

In my opinion, it will slowly become apparent that the way to harmonize our picture of human origins is to accept a substantial degree of archaic admixture in Africa. Such admixture cannot be detected directly, because there are no archaic genomes from Africa, and the hot climate throughout much of the continent may make preservation of DNA more difficult than in northern parts of Eurasia (where Neandertal and Denisovan individuals were from). Nor can it always be detected with LD-based methods, since LD decays exponentially and really old admixture is indistinguishable from an excess of mutation in a large population size. But, its acceptance will simultaneously solve the riddle of excess polymorphism in Africans, remove the need for an Out-of-Africa bottleneck of biblical proportions, and resolve the discrepancy between autosomal and uniparental evidence.

1 comment:

terryt said...

"really old admixture is indistinguishable from an excess of mutation in a large population size. But, its acceptance will simultaneously solve the riddle of excess polymorphism in Africans, remove the need for an Out-of-Africa bottleneck of biblical proportions, and resolve the discrepancy between autosomal and uniparental evidence".

And also explains a discrepancy note in Hawks' article:

"Until now, mitochondrial comparisons have been the strongest evidence in favor of a short time scale for the dispersal and differentiation of non-African peoples, within the past 70,000 y (17). Some recent attempts to examine the relationships of non-African populations using nuclear genome data have led to time scales in excess of 100,000 y"

There is surely really no reason why mt-DNA and Y-DNA dates should match aDNA, or each other. If we accept the wave theory of evolution we then easily accept that Y-and mt-DNA are just two more genes that have spread through the human species, independently in some cases.