Tuesday, March 25, 2008

Estimating Neandertal and modern human divergence

Had to review and analyze a scientific article and also compare it to our text. Def. shoulda chosen an easier read...

In their recent paper, Close correspondence between quantitative—and molecular—genetic divergence times for Neandertals and modern humans, Timothy Weaver, Charles Roseman, and Chris Stringer present a divergence time estimator to estimate when Neandertals and modern humans last shared a randomly mating common ancestor (split time) based upon the input of neutrally evolving morphological measurements. Recent research, including their own, has shown genetic drift (not natural selection) may have produced many of the cranial differences found between Neandertal and modern humans. If this research is correct, then they can accurately estimate population genetic parameters much in the same way as estimates are made from DNA sequences, the only difference is that they use modern human and Neandertal cranial measurements to make their estimates. Using advanced statistical formulas employed in evolutionary quantitative genetics and on microsatellites to develop their divergence time estimator, the researchers then apply this estimator to 37 cranial measurements collected on 2,524 modern humans from 30 globally distributed populations and 20 Neandertal specimens and calculate that Neandertals and modern humans split ≈311,000 (95% C.I.: 182,000 to 466,000) or 435,000 (95% C.I.: 308,000 to 592,000) years ago. The time-span of divergence that these researchers produce is relatively aligned with dates derived from ancient Neandertal and extant human DNA-sequence (see Noonan and colleagues 2006). This correspondence, according to the researchers, greatly strengthens the neutral divergence explanation for the cranial-facial differences found between Neandertals and modern humans and weakens the adaptionist explanations of diversifying natural selection. Lastly, they state that their work shows that there is no conflict between molecules and morphology.

The research methods employed in this paper are very complex and there are a number of variables and equations that I do not fully understand, so they are hard to critique. However, I would say that one of the difficulties of adapting DNA sequence statistics to morphological measurements, in order to determine the effects of genetic drift, is that there a lot more shifting variables to account for—namely population size and fluctuations. According to John Hawks, “a model of phenotypic evolution by genetic drift requires an assumption about the effective size of the population (Ne). Weaver et al. (2008) assume a model of "mutation-drift equilibrium." This is an assumption that the effective population size has not changed over time in the populations under consideration -- in this case, the Neandertal and human populations back at least as far as their common ancestor.” Weaver and colleagues set the effective population sizes of Neandertals and modern humans over the past thousand years at 2700 individuals. This number seems contradictory to and much smaller than existing evidence of actual population size and fluctuations. Such a small number would obviously emphasize the role that genetic drift played, but may be incorrect. Also, using recent phenotypic and genetic divergences of modern humans to 'calibrate their clock' of phenotypic evolution is questionable, since modern effective population sizes are much larger today, than they were during the Middle Pleistocene—which raises a serious question of comparability of phenotypic divergences between the two periods.


The article summarized above encompasses and synthesizes a number of topics and approaches referenced in our text book and advances and discredits some of the proposed explanations for variation (specifically cranial differences) found between Neandertal and modern humans. The researchers suggest that genetic drift, not natural selection, is largely responsible for many of the cranial differences found between Neandertals and modern humans. In chapter 13, our text book begins by elaborating on the environmental background of the Pleistocene period before it goes on to describe the Neandertal cranium. This is to emphasize the environmental impacts on the anatomical structure of the cranium, an adaptionist explanation. Our text references one opinion that “both the facial anatomy and the robust postcranial structure of Neandertals” according to Erik Trinkaus, of Washington University in St. Louis, are “adaptations to rigorous living in a cold climate” (p. 335). In an earlier paper (Weaver et al. 2006), the same researchers of the paper I referenced deemphasize these adaptionist explanations of natural selection and advance their own explanation: “isolation between Neandertal and modern human populations would have lead to cranial diversification by genetic drift (chance changes in the frequencies of alleles at genetic loci contributing to variation in cranial morphology).” Our book defines genetic drift as being “the random factor in evolution, it's a direct function of population size” (p.82). Our book also states that “drift occurs because the population is small” (p.82). Thus, the smaller the population size the greater genetic drift's influence is. The researchers do not, however, dismiss altogether the role played by natural selection on Neandertal and moder human craniums. They admit that the similarity in cranium size between Neandertal and modern humans was most likely heavily influenced by natural selection, but the subtle differences in shape found between the two craniums, they argue, are a result of genetic drift, not natural selection. In the paper the researchers also bring up neutral evolution, which is basically the idea that variation is random and unselective (in a Darwinian sense) and strongly influenced by genetic drift.

This paper, as has already been mentioned, uses a number of methods employed by population geneticists. In chapter 15 our book devotes a couple pages to this field. One name that pops up in both our text book and this article is the Hardy-Weinberg theory of genetic equilibrium. According to our text, “the Hardy-Weinberg theory of genetic equilibrium establishes a set of conditions in a population where no evolution occurs” (p.391). In the paper the researchers reference it to explain why there is a difference of a factor of two between the quantitative genetic and microsatellite forumals—basically, according to Hardy-Weinberg theory a gametic variance should be twice a zygotic variance. The Hardy-Weinberg theory is very useful because “by explicitly defining the genetic distribution that would be expected if no evolutionary change were occurring (that is, in equilibrium), we can compare the observed genetic distribution obtained in real human populations” (p.391). Thus, “if the observed frequencies differ from those of the expected model, we can then say that evolution is taking place at the locus in question. The alternative, of course, is that the observed and expected frequencies don't differ enough that we can confidently say evolution is occurring at a locus in a population” (p.391). By using and slightly adapting this strategy Weaver et al (2007) are able to determine whether genetic drift or natural selection are contributing more to the cranial differences found between Neandertals and modern humans.

The time span that the researcher's divergence time estimator produces is within the figures of recent research but slightly more recent than the dates suggested in our text from a 1997 report by Krings and colleagues who “hypothesized that the Neandertal lineage separated from our modern H. sapiens ancestors sometime between 690,000 and 550,000 ya” (p. 343). Before referencing this estimated divergence period our book explains the techniques used to produce these dates and determine the degree of genetic relatedness between species, which include: “extracting mitochondrial DNA (mtDNA), amplifying it through polymerase chain reaction, or PCR, and sequencing nucleotides in parts of the molecule” (p. 343). These techniques allow for more accurate estimates of time divergence which Weaver et al. (2008) can use to compare the accuracy of their own results to. Lastly, there has been a long and lively debate regarding the replacement of ancient Homo species with modern Homo. One hypothesis, the complete replacement model, which was developed by Bristish paleoanthropologists Christopher Stringer (one of the researchers in the paper) and Peter Andrews, “proposes that anatomically modern populations arose in Africa within the last 200,000 years and then migrated from Africa, completely replacing populations in Europe and Asia” (p.354). Another model, partial replacement, proposes that “some interbreeding occurred between emigrating Africans and resident premodern populations elsewhere” (p.356). In other words, the partial replacement model assumes that “no speciation event occurred, and all these hominids should be considered members of H. sapiens (p.356). Then, lastly, there's the regional continuity model, associated with Milford Wolpoff, which suggests that “local populations—not all, of course—in Europe, Asia, and Africa continued their indigenous evolutionary development from premodern Middle Pleistocene forms to anatomically modern humans” (p.356). The last two models are questionable for a number of reasons and the findings in the Weaver et al. (2008) paper support Stringer and Andrew's replacement model, since they show that phenotypic differences in cranial measurements can be used to trace back to the time in which these two species diverged.


The article I reviewed at first seemed extremely esoteric and beyond comprehension, and so it often felt like I was reading something that was written in an entirely different language. It required me to learn quite a bit about quantitative genetics, microsatellites, neutral evolution, genetic drift, and mutation-drift equilibrium—terms which I had no knowledge of, at all, before. Once I familiarized myself with these terms and the methods employed in the paper I was able to get a pretty good grip on the substance of what they are saying.

The researchers are suggesting that natural selection is not the only or even strongest influence on diversity (or differences found between species) and argue that genetic drift, in the case of cranial differences found between Neandertal and modern humans, appears to have played a much stronger role. More importantly, by using morphological cranium measurements and plugging it into their formulas to produce similar or overlapping dates with the ones derived from DNA sequencing Weaver et al. (2008) build a potential bridge between the molecular and morphology camps. The only problem that I would raise with their paper is my concern that their formulas do not accurately address true population sizes and fluctuations. Based on the evidence we have it seems Neandertal and modern human population sizes waxed and waned at incongruent intervals. If the population sizes dwindled to small numbers diversity diminishes and deleterious traits begin to emerge. The researchers assume the population size to be 2700 individuals, which as John Hawks puts it “is an astounding assumption.” A small effective population size, such as the one the researchers select, means that rapid evolution will occur by genetic drift. This small size, however, is disputed by other evidence. As John Hawks again points out:

“most other sets of genetic data indicate a long-term effective size of at least 10,000 for human populations -- four times larger than assumed in this study. All things being equal, this means that the rate of phenotypic evolution by genetic drift should be four times slower than assumed by Weaver et al. (2008). Some of this difference between real and assumed effective sizes may be washed out by their process of calibration -- their equations involve several unknowns that must be simultaneously estimated, and give a lot of wiggle-room to the results.”

The problem with statistical formulas that include these unknown variables is that they can be adjusted such that the results can be fine-tuned and any phenotypic difference can thus look like genetic drift.

Lastly, in their formula they use 37 cranial measurements collected on 2,524 modern humans from 30 globally distributed populations. Modern human populations have dramatically increased in size, so whether one can use these measurements to accurately measure the differences between modern human craniums and Neandertal to track their divergence seems to me questionable. Overall I thought the article was very interesting and I think the researchers are usefully adopting and synthesizing numerous methods from various fields which bring molecular research and morphology closer together, producing a much clearer and accurate picture of our past.


Weaver, T.D., Roseman, C.C., Stringer, C.B. (2008). Close correspondence between quantitative- and molecular-genetic divergence times for Neandertals and modern humans. Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0709079105

No comments: