You may already know that I am highly critical of the often too generalized assumption that molecular clock based age estimates have any strong validity. Way too many people just assume that if geneticists say that certain lineage has this or that age, that is almost the equivalent of C14 datations, which are quite accurate and repeatedly proven. That is not just not true at all. Molecular clock theory is only very weakly proven and critical variables such as the exact mutation rates, the effective population sizes involved, the influence of factors such as weak selection and the validity of the mathematic constructs used to evaluate these ages are all too poorly understood in fact.
So I want to raise your attention on the recent spat of research papers on this matter, with contradictory results but all interesting to read in any case. They are:
- Pedro Soares et al., Correcting for Purifying Selection: An Improved Human Mitochondrial Molecular Clock. AJHG, 2009.
- Eva-Liis Loogväli et al., Explaining the Imperfection of the Molecular Clock of Hominid Mitochondria. PLoS ONE 2009.
- Philip Endicott et al., Evaluating the mitochondrial timescale of human evolution. Trends in Genetics, 2009.
Each of them has as co-author one or several "big names" on the field of human population genetics. Soares' paper is the oldest by some months and hence cited by the others, albeit critically, and is backed by Oppenheimer, McAulay and Richards. Loogväli's paper has Thomas Kivisild as co-author, while Endicott's research is co-signed by Mait Metspalu.
As said before, they each reach to different conclusions though they are complex enough that I'd shy a bit of analyzing them in depth myself here. Suffice to say, as illustrative example, that Soares concludes that mtDNA N is older than M (and that South Asian M is more recent than East Asian M - an artifact of their method in my opinion), while Loogväli concludes exactly the opposite, assigning to M an estimate age of c. 77,000 years and to N of 68,000 years. Meanwhile Endicott is much more cautious and makes instead a critical review of the most important literature on the matter, along with the archaeological data in a very comprehensive manner. However he seems persuaded that, contrary to Oppenheimer's catastrophist hypothesis, the deep origin of Eurasians is in South Asia.
In fact, while finding interesting the other papers too, more Loogväli's than Soares' (who seems to fall in too much in the errors generated by statistical artifacts and relies heavily on similarly controversial Pan-Homo divergence estimates), I am surely more in agreement with Endicott's caution in this highly complex and controversial matter. Hence I will reproduce here his concluding remarks (adding my emphasis in bold type), demanding greater rigor:
Further research is needed to improve our confidence in molecular estimates of human evolutionary timescales. First, the most reliable calibrations within the human tree need to be identified. For mitochondrial DNA, this depends on finding well-defined haplogroups that can be precisely associated with dated palaeoanthropological evidence [17]. Second, the variation in observed rates across different timescales needs to be accurately quantified [16–18]. Third, these patterns of rate variation need to be investigated for nuclear data, including the Y-chromosome and short tandem repeats.
The chief recommendation arising from the current state of knowledge in the field is for a movement away from reliance on the human-chimpanzee calibration; instead, calibrations within the human tree are preferred (but see [14]). There are several recent examples of estimates made using archaeological calibrations [15–17,35], extending the efforts of earlier authors [3,60]. Considering recent advances in phylogenetic methodology, there is now a compelling motivation to employ statistical models that take into account rate heterogeneity among sites and among lineages, that correct for multiple substitutions (saturation), and that incorporate directly the uncertainty in the ages of calibrations used. Some methods also allow the statistical evaluation of competing demographic models, which can have an important influence on estimates of rates and timescales [17,23].
Important update (Jan-1-2010):
I just realized I had missed what seems to be a very important paper on the debate on the molecular clock applied specifically to human mtDNA also published in 2009:
Brenna M. Henn et al., Characterizing the Time Dependency of Human Mitochondrial DNA Mutation Rate Estimates. Oxford Journals - Molecular Biology and Evolution, 2009 (open access).
I am still finishing it but it seems a most crucial and balanced analysis on the debate (within the molecular clock hypothesis) between the phylogenetic and pedigree mutation rates' camps.
As far as I can tell, Henn and colleagues largely respond (several months in advance) to some of the recommendations made by Endicott in his paper, using in fact archaeological calibrations to estimate the reality of the mutation rates. They conclude that while pedigree rates (decreased by 33% to fit with the realistic generation span of 30 years) are almost identical to reality for the last 5000 years, they are totally unrelated with the actual behaviour of lineages before c. 15,000 BP, when the phylogenetic rates become the only reliable ones. The lapse between these two dates behaves anomalously (rather sudden but irregular change from one model to the other) leaving a gray transitional zone, difficult to work with.
Notably (as some of the hottest debates are about the age of mtDNA haplogroup H), they conclude that H3 probably has an age of c. 18,000 BP, while H1 could be even quite older. This is totally consistent with the fact that North African H (including specially these two subclades) appears derived from Iberian mtDNA and with this haplogroup having been found in aboundance in ancient pre-Neolithic DNA in both Morocco and Portugal.