Razib Khan One-stop-shopping for all of my content

December 18, 2012

Unveiling the genealogical lattice

To understand nature in all its complexity we have to cut down the riotous variety down to size. For ease of comprehension we formalize with math, verbalize with analogies, and visualize with representations. These approximations of reality are not reality, but when we look through the glass darkly they give us filaments of essential insight. Dalton’s model of the atom is false in important details (e.g., fundamental particles turn out to be divisible into quarks), but it still has conceptual utility.

Likewise, the phylogenetic trees popularized by L. L. Cavalli-Sforza in The History and Geography of Human Genes are still useful in understanding the shape of the human demographic past. But it seems that the bifurcating model of the tree must now be strongly tinted by the shades of reticulation. In a stylized sense inter-specific phylogenies, which assume the approximate truth of the biological species concept (i.e., little gene flow across lineages), mislead us when we think of the phylogeny of species on the microevolutionary scale of population genetics. On an intra-specific scale gene flow is not just a nuisance parameter in the model, it is an essential phenomenon which must be accommodated into the framework.

This is on my mind because of the emergence of packages such as TreeMix and AdmixTools. Using software such as these on the numerous public data sets allows one to perceive the reality of admixture, and overlay lateral gene flow upon the tree as a natural expectation. But perhaps a deeper result is the character of the tree itself is torn asunder. The figure above is from a new paper, Efficient moment-based inference of admixture parameters and sources of gene flow, which debuts MixMapper. The authors bring a lot of mathematical heft to their exposition, and I can’t say I follow all of it (though some of the details are very similar to Pickrell et al.’s). But in short it seems that in comparison to TreeMix MixMapper allows for more powerful inference of a narrower set of populations, selected for exploring very specific questions. In contrast, TreeMix explores the whole landscape with minimal supervision. Having used the latter I can testify that that is true.

The big result from MixMapper is that it extends the result of Patterson et al., and confirms that modern Europeans seem to be an admixture between a “north Eurasian” population, and a vague “west Eurasian” population. Importantly, they find evidence of admixture in Sardinians, which implies that Patterson et al.’s original were not sensitive to admixture in putative reference populations (note that Patterson is a coauthor on this paper as well). The rub, as noted in the paper, is that it is difficult to estimate admixture when you don’t have “pure” ancestral reference populations. And yet here the takeaway for me is that we may need to rethink our whole conception of pure ancestral populations, and imagine a human phylogenetic tree as a series of lattices in eternal flux, with admixed nodes periodically expanding so as to generate the artifice of a diversifying tree. The closer we look, the more likely that it seems that most of the populations which have undergone demographic expansion in the past 10,000 years are also the products of admixture. Any story of the past 10,000 years, and likely the past 100,000 years, must give space at the center of the narrative arc lateral gene flow across populations.

Cite: arXiv:1212.2555 [q-bio.PE]

October 18, 2012

Most mtDNA lineages expanded before the Neolithic?

Filed under: Human Genetics,Human Genomics,phylogenetics — Razib Khan @ 8:00 pm

A new short communication in Scientific Reports suggests that most demographic expansion as ascertained using mtDNA occurred before the Neolithic. MtDNA analysis of global populations support that major population expansions began before Neolithic Time:

Agriculture resulted in extensive population growths and human activities. However, whether major human expansions started after Neolithic Time still remained controversial. With the benefit of 1000 Genome Project, we were able to analyze a total of 910 samples from 11 populations in Africa, Europe and Americas. From these random samples, we identified the expansion lineages and reconstructed the historical demographic variations. In all the three continents, we found that most major lineage expansions (11 out of 15 star lineages in Africa, all autochthonous lineages in Europe and America) coalesced before the first appearance of agriculture. Furthermore, major population expansions were estimated after Last Glacial Maximum but before Neolithic Time, also corresponding to the result of major lineage expansions. Considering results in current and previous study, global mtDNA evidence showed that rising temperature after Last Glacial Maximum offered amiable environments and might be the most important factor for prehistorical human expansions.

A good aspect of this result is that they used whole mtDNA sequences. Why use short ...

October 7, 2012

The end of phylogenetic controversies

Filed under: phylogenetics — Razib Khan @ 10:10 pm

I put up kind of a ridiculous title. But I do hope that at some point in the near future we’ll have some of the same flavor of debates on the macroevolutionary time scale that we have on the human microevolutionary time scale. There’ll be a surfeit of sequence at nearly every node of interest on the tree of life, and computational power galore devoted to analyzing variation and reconstructing any phylogeny we can conceive of. To be fair, one could argue we aren’t there even with human phylogenetics either. But it is rather strange we’re debating the origin of mammals and the nature of the lineage’s phylogenetic tree at this time. This is the kind of thing that I hope a more robust and assertive molecular phylogenetics can resolve (and paleontology as well, but I’m not up on the latest in computational analysis of morphological characters).

In any case, here’s a cool manuscript up at arXiv, A phylogenomic perspective on the radiation of ray-finned fishes based upon targeted sequencing of ultraconserved elements:

Ray-finned fishes constitute the dominant radiation of vertebrates with over 30,000 species. Although molecular phylogenetics has begun to disentangle major evolutionary relationships within this vast section of ...

August 19, 2012

On phylogenetic instrumentalism

Filed under: Genomics,Human Evolution,Human Genetics,Human Genomics,phylogenetics — Razib Khan @ 4:40 pm

ADMIXTURE and STRUCTURE tests aren’t formal mixture tests. Yes! In fact, in the “open science” community this issue is repeated over and over and over, because people routinely get confused (our audience does not consist of population geneticists and phylogeneticists by and large). So sometimes it is necessary to lay it out in detail as in the post above. The key point to always remember is that population genetic & phylogenetic statistics and visualizations are a reduction and summary of reality in human palatable form. They tell us something, but they do not tell us everything. A common issue is that for purposes of mental digestion it is useful to label ancestral elements “European,” or on PCA refer to a “European-Asian” cline, as if the population genetic abstractions themselves are the measure of what European or Asian is. But European and Asian are themselves human constructions, and subject to debate (e.g., do Turks count as Europeans? Indians as Asians?) The population genetic statistics are not themselves subjective, but the meanings we give them are.

Let’s illustrate this with a concrete example. The Cape Coloured population of South Africa is a compound of Khoisan, Bantu, South Asian, Southeast Asian, and ...

August 10, 2012

On the limits of feathery trees

Filed under: Genetics,phylogenetics — Razib Khan @ 11:53 pm

The NJ tree is from Genome-Wide Analysis in Brazilian Xavante Indians Reveals Low Degree of Admixture. It’s a visualization of a genetic distance matrix. Am I strange, or do these sorts of trees really leave a lot to be desired in terms of actually getting across any extra information beyond a table?

July 12, 2011

Bearish wisdom!

Filed under: Brown Bears,Genetics,Genomics,Irish Bears,phylogenetics,Polar Bears — Razib Khan @ 12:11 am

ResearchBlogging.orgA few years ago a paper came out which suggested that the brown bears of the ABC Islands of southeastern Alaska were more closely related to polar bears than they were to other brown bears. More precisely, polar bears and ABC brown bears formed a distinct clade set apart from other brown bears, so that the class “brown bear” was not monophyletic. This meant that all the descendants of the hypothetical ancestral lineage of brown bears are not brown bears. Like reptiles, brown bears may then be paraphyletic. If this is correct polar bears can be thought of as a derived and specialized lineage of brown bears, despite all their morphological differences.

This is not just systematic arcana. The phylogenetic relationships of species has important implications for their conservation status, something all the more salient due to changes in the arctic habitat of the polar bear.

But there is a catch with the science: it focuses on mitochondrial lineages. In other words, the matriline, the female line of descent. There are technical reasons for this, primarily having to do with the tractability of ...

April 27, 2011

The continuing tangling of the human tree

ResearchBlogging.orgLast summer I made a thoughtless and silly error in relation to a model of human population history when asked by a reader the question: “which population is most distantly related to Africans?” I contended that all non-African populations are equally distant. This is obviously wrong on the face of it if you look at any genetic distance measures. West Eurasians, even those without recent Sub-Saharan African admixture (e.g., North Europeans) are closer than East Eurasians, who are often closer than Oceanians and Amerindians. One explanation I offered is that these latter groups were subject to greater genetic drift through a series of population bottlenecks. In this framework the number of generations until the last common ancestor with Sub-Saharan Africans for all groups outside of Africa should be about the same, but due to evolutionary factors such as more extreme genetic drift or different selective pressures some non-African groups had diverged more from Africans than others in terms of their genetic state. In other words, the most genetically divergent groups in relation to Africans did not diverge any earlier, but simply diverged more ...

April 24, 2011

Beware of one tree!

Filed under: Anthroplogy,Human Genetics,phylogenetics — Razib Khan @ 4:59 pm

Sometimes in the comments of this weblog people get into heated disagreements about one figure and its proper interpretation. I don’t get much involved most of the time because different visualization techniques often differ on the margin, so getting obsessed with minor details is a fool’s errand. For example, in the paper I reviewed below there was a neighbor-joining phylogenetic tree. Take a look. The length of the branches are proportional to genetic distance.

December 23, 2010

The paradigm is dead, long live the paradigm!

Mitochondrial DNA and human evolution:

Mitochondrial DNA from 147 people, drawn from five geographic populations have been analysed by restriction mapping. All these mitochondrial DMAs stem from one woman who is postulated to have lived ab7out 200,000 years ago, probably in Africa. All the populations examined except the African population have multiple origins, implying that each area was colonised repeatedly

And so was published in the year 1987 the paper which established in the public’s mind the idea of mitochondrial Eve, which gave rise to a famous cover photo in Newsweek. This also led to the Children of Eve episode on the PBS documentary NOVA. Here is the summary:

NOVA examines a controversial theory that traces our ancestry to a small group of women living in Africa 300,000 years ago.

As Milford Wolpoff has complained it is probably accurate to characterize the documentary as not particularly “fair & balanced.” Mitochondrial Eve may have been controversial, and subsequently plagued by issues of molecular clock calibration as well as spurious interpretations of the cladograms, but the tide of history was on its side, and PBS was telling that story. And the story was not just the primary science, rather, one had to understand the controversy in light of the debates among paleontologists and between paleontologists and molecular biologists. A group of researchers, spearheaded by Chris Stringer argued for the recent origin of modern humans from Africa on the basis of fossils alone. They were challenged by an established school of multiregionalists who argued for deeper roots of modern human populations, which derived from local hominins which diversified after the the migration of H. erectus out of Africa. The argument of the multiregionalists was that selective sweeps across the full range of the human populations gave rise gradually to modern humanity as we know it, a compound of specific ancient local features and trans-population characters which unified us into a broader whole. Stringer and company presented a simpler model where anatomically modern human being arose ~200,000 years ago in Africa, and subsequently expanded to other parts of the world, by and large replacing the local hominin populations. In the multiregionalist telling Neandertals became human beings, while Out of Africa would imply that Neandertals were replaced by human beings.

ResearchBlogging.orgInto this tendentious landscape of bones stepped the molecular biologists. The critical figure here is Allan Wilson, who in the 1970s argued forcefully from molecular clock evidence for a more recent separation of the human and ape lineage than paleontologists had favored. By the 1980s the paleontologists had generally conceded that Wilson et al. were correct. After this victory he put forward the mitochondrial Eve theory with his student Rebecca Cann. Here Wilson was getting involved with an argument about paleontology. From all the material I’ve read Wilson and Cann were confident that their techniques were superior to old fashioned analysis of fossils, a method which Wolpoff defended vociferously on NOVA. People who were not invested in recent human origins often did not know what to make of the debate. To give you a flavor of what was going on in the late 1980s, here’s Richard Leakey in Origins Reconsidered: In Search of What Makes Us Human:

……In the 1970s, I have been more reluctnant than most to accept Wilson and Sarich’s genetic evident in favor of a recent (five million years ago) origin of hominids, so I thought this would be a chance to redress the balance. In thecourse of my talk I mentioned the mitochondrial DNA evidence and indicated that “I was ready to be persuaded by it.” Surrounded as I was by molecular biologists and geneticists, I imagined it would be a wise think to do, and scientifically proper too.

I was therefore more than a little surprised when, in the bar after my talk, several participants, including the conference organizer, Stepehen O’Brien, cornered me and said, “You don’t have to swallow the Mitochondrial Eve line. We don’t.” Steve and his friends proceeded to tell me why they thought the Eve hypothesis was incorrect…Wilson may have miscalculated the rate of the mitochondrial clock, older mitochondria may have been lost by chance, promoted perhaps by occasional crashes in local pouplation size, natural selection may have favored some recent evolved mitochondrial variant, this eliminating the older lineages. Any of these possibilites might erroneously lave the impression of a recently emerged population….

…In February 1990, Milford and a half a dozen like-minded colleagues organized a session at the annual gathering of the American Association for the Advancement of Science, in New Orleans, the goal of which was to “nail this Mitochondrial Eve nonsense.” Speaker after speaker argued for evidence in support of regional continuity and against localized speciation; for alternative interpretations…It was a powerful presentation, and gathered a lot of press, with headlines like “Scientists Attack ‘Eve’ Theory of Human Evolution” and “Man Does not Owe Everything to Eve, Latest Finding Says.” Chris Stringer, who was speaking at a different session of the meeting, described the anti-Eve seminar as “high-powered salesmenship.” One of Milford’s assault team, David Frayer of the University of Kansas, summarized the deep reaction to Wilson’s work: “Fossils are the real evidence.”

In the 1990s Wolpoff came out with a book, Race and Human Evolution: A Fatal Attraction. It outlined a multiregionalist framework for the origin of modern humans, and also presented a wide ranging review of human paleoanthropology past to present, and, to my eyes made the case that the multiregionalists were on the “right side of history.” I was, and remain, a natural history nerd. Especially a natural history nerd of the human species. I devoured books on the topic in the 1980s and 1990s, and saw the slow shift away from multiregionalism toward an Out of Africa model as the orthodoxy, as transmitted by scientific journalists. As I did not have any horse in the race, it was not a matter of concern either way for me, but, I did observe that the disagreements were personal and sometimes politicized. Race and Human Evolution seems to have been written in part to debunk the idea that multiregionalism gave succor to racism. Rather, Wolpoff inverted the narrative, presenting Out of Africa models as genocidal and exterminationist, in contrast to his model of human populations gliding toward sapiency together through gene flow.

The flip side of course is that many people presented Out of Africa as anti-racist par excellence. Anatomically modern humans were portrayed as the latter day Julius Caesar’s of the hominin world. They came, they saw, and they conquered. The chasm between humans and non-humans may have been wide, but the more appealing aspect of the Out of Africa model is that we were the new kids on the block. All non-African humans derived from Africans, who were the reservoirs of our species’ genetic diversity. The dovetailing of implications of the model with the egalitarian ethos of the age was natural. Here is Pat Shipman in 2003, We Are All Africans:

I don’t expect that the subscribers of the Multiregional hypothesis will be waving a white flag of surrender, although they have lost the great majority of their supporters. At least one of the theory’s most ardent proponents, Wolpoff, is still steadfast in defense of the hypothesis he has so long espoused. While it remains possible that new findings will shift the balance in favor of the Multiregional viewpoint, the consilience of such evidence creates a powerful testament. It would take many new fossils and many new genetic studies to resculpt this intellectual landscape.

By and large the arguments which Shipman lays out were persuasive to someone like me who didn’t know much about bones & stones. Though even I knew of some instances of possible continuity, the mtDNA, Y chromosomal lineages, and autosomal results, did seem to roughly line up appropriately. In the battle between paleoanthropologists who saw continuity in the fossils and those who did not, it seemed reasonable to at the time to give the “tiebreaker” to the geneticists who were generating inferences consistent with Out of Africa.


With all that said, it has to be stated that paleoanthropologists such as Chris Stringer did not hold necessarily to total replacement of non-Africans. Total replacement may have been the case, but quite often they did qualify that there may have been some admixture and assimilation with the pre-modern substrate. But the paucity of the genetic data pointing to interbreeding between distant lineages (as opposed to a very recent exclusive common ancestry), especially once the Neandertal mtDNA was shown to be an outgroup, seems to have pushed people to the model where modern humans were an entirely different beast which simply wouldn’t have deigned to to have intercourse with the creatures of yore. In The Dawn of Human Culture the paleoanthropologist Richard Klein lays out a scholarly and measured argument for what is close to a maximalist case for the unique and distinctive nature of modern neo-African humanity:

……the simplest and most economic explanation for the “dawn” is that it stemmed from a fortuitous mutation that promoted the fully modern human brain….an acknowledged genetic link between anatomy and behavior in yet earlier people persisted until the emergence of fully modern ones and that the postulated genetic change 50,000 years ago fostered the uniquely modern ability to adapt to a remarkable range of natural and social circumstances with little or no physiological change.

Arguably, the last key neural change promoted the modern capacity for rapidly spoken phonemic language, or for what anthropologists Duane Quiatt and Richard Milo have called “a fully vocal language, phonemicized, syntactical, and infintely open and productive.”

Wolpoff was on to something. Even if the original Out of Africa proponents did not mean to do so, there was a tendency to remove “higher faculties” from the suite of capabilities of the evolutionary “dead ends.” We were H. sapiens sapiens. If we deigned to allow Neandertals to be a branch of our own species, their subspecies was distinctive. They were less than we in the ways in which modern humans were exceptional, and universal.

This orthodoxy probably resulted in a positive feedback loop for the educated public, in which I include myself. The more the Out of Africa model of neo-African human exceptionalism settled into the received wisdom, the more animalized Neandertals and other human lineages became. Naturally a multiregionalist model of continuity became distasteful, because continuity implied a connection between modern humans and subhumans. The fact that the largest cranial capacities in the whole human lineage were sported by Neandertals became a counterintuitive fact, which just went to show that it was quality, not quantity.

When I was a freshman at university I took a biological anthropology course. The instructor threw out a question to the class. He noted that some paleoanthropologists observed a continuity between the skulls of Australian Aborigines and some Southeast Asian erectine populations. Australian Aborigines are a very robust people, and have been less affected by the trend toward gracility which has been the norm over the past 10,000 years for most human populations. In any case, the instructor asked for a show of hands whether such a possibility should even be discussed openly. The solid majority of the class rejected an open discussion. When asked by the instructor why, many of the students who rejected an examination of the thesis argued that such a possibility opened the path to de-humanization, oppression, and was politically too sensitive. Milford Wolpoff had obviously lost the propaganda war. The students did not consider the possibility of multiregionalism where all human populations exhibited continuity, rather, they assumed that continuity hypothesized for Australian Aborigines was specific to them, and so would associate that population with the less human branches of the hominin tree.

Science is a human cultural endeavour. It is about something real, something objective, but we do look through the glass somewhat darkly. The acceptance or rejection of models are contingent upon correspondence to reality and precision of prediction. But the rise and fall of models, and the rate of their rise and fall, may be subject to cultural dynamics. In The Price of Altruism Oren Harman shows how the cultures of Russia and Britain shaped how they viewed the social implications of evolutionary biology. Similarly, Newtonian mechanics and Darwinian evolution may have been retarded in their initial acceptance in France due to reasons of language and national chauvinism.

Not only do scientific theories have to swim through the waters of suspicion and incomprehension across societies, but they also have to overcome the inevitable confounding of their natural inferences with normative ones. Newtonian mechanics, relativity, and quantum mechanics, have all had many peculiar and surprising downstream social consequences. The line made between these physical theories and models and sociology, epistemology, and spirituality, would likely have surprised their originators (OK, perhaps not Isaac Newton). But the human imagination is fertile, and many cognitive anthropologists argue that the connections and analogies that we make, in addition to our promiscuous pattern recognition, gives rise the baroque and baffling complexity that is culture.

By the mid-2000s the paradigm of Out of Africa had crystallized to such a point that even the fossils purportedly betrayed the multiregionalists. In Bones, Stones and Molecules: “Out of Africa” and Human Origins the authors made the case that the fossil record, and its pattern of variation, complemented the molecular record. That is, Chris Stringer was right. Other more computationally intensive analyses of morphological variation reportedly tended to support an Out of Africa model.

And yet just as Out of Africa seemed to have cleared the field, pointers in the other direction were bubbling up out of genomics and genetics. In 2006 Bruce Lahn at the University of Chicago published Evidence that the adaptive allele of the brain size gene microcephalin introgressed into Homo sapiens from an archaic Homo lineage. Nevertheless several years later there seems to have been no wide support for this hypothesis. For eample, No evidence of a Neanderthal contribution to modern human diversity. But there were other papers nonetheless. Deep Haplotype Divergence and Long-Range Linkage Disequilibrium at Xp21.1 Provide Evidence That Humans Descend From a Structured Ancestral Population. Genomics refutes an exclusively African origin of humans. Granted, this was a minority perspective. For the first few years the Neandertal genome project did not seem to support any admixture either. I saw Svante Paabo speak in late 20008, and he was absolutely unequivocal. No sign of admixture. Period.

But the equilibrium of scientific orthodoxy is not eternally robust to a hard exogenous shock of falsification. Yes, some scientists remain obstinate in the face of overwhelming evidence. One could argue Milford Wolpoff could be numbered amongst these. Fred Hoyle certainly was. But the tide turns. In the fall of 2009 Svante Paabo seemed to be far less unequivocal about the issue of admixture. Then, in the spring of 2010:

A test of the New Mexico team’s proposals may come soon. Svante Pääbo and colleagues at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, announced early last year that they had finished sequencing a first draft of the Neanderthal genome, and they are expected to publish their work in the near future. Pääbo’s earlier studies on components of Neanderthal genomes largely ruled out interbreeding, but they were not based on more comprehensive analyses of the complete genome.

Linda Vigilant, an anthropologist at the Planck Institute, found Joyce’s talk a convincing answer to “subtle deviations” noticed in genetic variation in the Pacific region.

“This information is really helpful,” says Vigilant. “And it’s cool.”

By this point, in April of 2010, some graduate students who were not involved in the project itself had seen hard copy drafts of the Neandertal admixture paper. Word was spreading. I already knew of its likely probability, which resulted in me turning on Google Alerts (which got me in trouble for “breaking embargo” on an embargo which I was never privy to). The hammer-blows against the old tried & true orthodoxy in 2010 were ripening throughout the year, and many people were “in the know.” In the age of transparency it is interesting that science naturally has a culture of some secrecy. Who wants to be scooped? But how sustainable is this really over the long term?

To use a religious analogy which some may find offensive, this was an instance where the heretics were once the high priests of the faith. The media reports from last spring made it clear that most of the principals involved did not initially believe that admixture had occurred. Rather, they assumed that the results they were getting were anomalies. Science is influenced by culture, but ultimately nature remains the final arbiter. The truth is what it is, and honest men and women give it its due.

At this point you presumably know the score. Ancient DNA is a powerful judge and jury. It seems that the evidence for Neandertal admixture is already modifying the conventional Out of Africa narrative. But, it has to be admitted that Out of Africa is predominantly correct. The vast majority of our total genome content seems to be traceable to African populations within the last ~100,000 years. An older model of deep rooted lineages only periodically punctuated by selective sweeps which maintain species cohesiveness is not tenable. Phyletic gradualism seems implausible in light of the genetic evidence. Here is Wolpoff (and his wife, Rachel Caspari) in Race and Human Evolution:

We agree a punctuated evolutionary pattern best describes the evolutionary histories of many phyletic groups, including, we think, the earlier and much longer part of human prehistory when humans were only another African primate species. But we believe punctuated equilibrium does not reflect what happened to humans in the later part of human evolution as they became successful colonizers and when there was no macroevolutionary change. As we read the fossil record, there is no evidence of speciation events in the recent past; in fact, there is strong evidence against them. But the Eve interpretation promised to support a punctuated model for later human evolution that was denied by interpretations of the fossil evidence such as ours.

I’m not knowledgeable enough to know what would qualify as a “macroevolutionary change.” But the ‘Great Leap Forward’ seems a plausible candidate. Whatever the details, between 200 and 10 thousand years ago, there does seem to have been a series of rapid expansions of the human range and capacity for innovation. Sometime different was in the air. I do not know the nuance of Milford Wolpoff’s thinking. The most recent data do seem to refute the contention that all ancestry but the Out of African is trivial. But, they also seem to be broadly in line with the peculiarity, almost revolutionary character, of the changes in the human lineage over the past 200,000 years. Convergent patterns of morphological and genetic variation which seem to root back to an African base indicate that Chris Stringer and Allan Wilson had properly characterized a major first order dynamic in recent human prehistory. But now we move into the second and third orders. The rough paradigm is getting sculpted into something with more verisimilitude when judged against the diversity and peculiarity of nature.

Let’s jump to the paper. The main course. Genetic history of an archaic hominin group from Denisova Cave in Siberia:

Using DNA extracted from a finger bone found in Denisova Cave in southern Siberia, we have sequenced the genome of an archaic hominin to about 1.9-fold coverage. This individual is from a group that shares a common origin with Neanderthals. This population was not involved in the putative gene flow from Neanderthals into Eurasians; however, the data suggest that it contributed 4–6% of its genetic material to the genomes of present-day Melanesians. We designate this hominin population ‘Denisovans’ and suggest that it may have been widespread in Asia during the Late Pleistocene epoch. A tooth found in Denisova Cave carries a mitochondrial genome highly similar to that of the finger bone. This tooth shares no derived morphological features with Neanderthals or modern humans, further indicating that Denisovans have an evolutionary history distinct from Neanderthals and modern humans.

John Hawks has covered a great deal of ground in his FAQ. In particular, he has a gestalt understanding of the fossil record so he can run “quick & dirty” checks on some of their assertions. He notes:

What the paper doesn’t point out is that there are Upper Paleolithic specimens that equal or exceed this tooth in size. For example, the measured length and breadth of an upper second molar from Oase, Romania, are larger than this specimen, and the third molar (in the crypt) of that specimen is yet larger. There is an Upper Paleolithic-associated molar from Turkey which is also exceedingly large.

I don’t take that as a sign of relationship between this specimen and early Upper Paleolithic people — even though these are some of the earliest. It is another sign of how non-diagnostic this tooth actually is. I would say that in the absence of genetic information, we’d be looking at these remains as likely early Upper Paleolithic people, and accentuating these similarities.

People interpret information in light of their background priors. Now that we know what we did not, it may behoove us to go back and double check we may once have dismissed. Consider this paper from 2006, Archaic admixture in the human genome:

One of the enduring questions in the evolution of our species surrounds the fate of ‘archaic’ forms of Homo. Did Neanderthals go extinct without interbreeding with modern humans 25–40 thousand years ago or are their genes present among modern-day Europeans? Recent work suggests that Neanderthals and an as yet unidentified archaic African population contributed to at least 5% of the modern European and West African gene pools, respectively. Extensive sequencing of Neanderthal and other archaic human nuclear DNA has the potential to answer this question definitively within the next few years.

5% is a nice round number. They could have lucked upon it, but the first author continued to plunge onward in 2009, generating models of archaic admixture. How fruitful would this be? Here is Sarah Tishkoff in December of 2009:

…Sarah Tishkoff, a geneticist at the University of Pennsylvania, agrees, adding that, after all, every population has a strong selective pressure for intelligence, the better to succeed in its respective environment. As far as consorting with Neanderthals, Tishkoff dismisses that notion as pure speculation: “I don’t know of any evidence for that.”

I suspect that Sarah Tishkoff’s opinion would have been common among most scholars of human evolution in late 2009 (though I suspect those who were Facebook friends with people in Svante Paabo’s lab perhaps not). To be fair to Tishkoff, she had no compunction about accepting Neandertal admixture six months later when presented with evidence. She even added that “…it is possible that interbreeding introduced traits into a few human populations.”

In regards to the paper, the top line is rather clear in the three figures in the article proper. I’ve reformatted them a bit below:

Top left: a phylogenetic tree which shows the total genome relationship of various human lineages. Extant modern humans represent one clade. The Denisovans and Neandertals another. In other words, the last common ancestral population of Denisovans and Neandertals is shallower in time than the last common ancestral population of neo-Africans and the Denisovans and Neandertals. All the Neandertals also are very closely related, at least when graded on this particular curve. The Denisovans are outgroups to them, just as the San are outgroups to other humans. The French are an outgroup to the Han and Papuans, though just barely. This sort of relationship is naturally why I cast a skeptical eye to arguments of the common ancestry of French and Han 20,000 years ago when we know that the Papuans settled their island 45,000 years ago.

Top right: a PCA where HGDP populations are projected onto the two largest components of variation which shake out of a data set of a chimpanzee, Denisovan, and Neandertal. In other words, the ones deciding the rules of the game here are chimps and the two archaic Eurasian populations. Humans are constrained onto the genetic variation space of non-/pre-humans. So the position of the humans tells you how they relate to the genetic variation of the Denisovans, Neandertals, and chimpanzees. The Eurasicans, Eurasians + Amerindians, form a relatively tight cluster, apart from Africans. If non-Africans have some Neandertal admixture, this is reasonable. But interestingly the Melanesian groups stand apart as well. And, Papuans and Bougainville Islanders are also distinctive. The latter are shifted toward Eurasicans. Why? Probably because they have a minor, but significant, Austronesian ancestral component which the Papuans lack.

They estimate that 2.5% of the genes of Eurasicans and Oceanians is of Neandertal origin. And, a further 5% of the Melanesian genome is of Denisovan origin. So Melanesians are 92.5% neo-African. Eurasicans are 97.25% neo-African. At most.

Bottom: the last shows a stylized demographic model. Step 1, humans leave Africa. The neo-Africans interbreed with southwest Asian Neandertals. Step 2, the paleo-Eurasians push east, and some encounter the Denisovans, eventually reaching Sahul ~45,000 years ago.

Some people have asked me about the Denisovan in Polynesians and Australian Aborigines. Since Polynesians are ~20% Melanesian, they should have a fraction diluted appropriately. As for Australians, if they are only recently distinguished from the peoples of Papua because of rising sea levels I assume that they should carry the same fraction of Denisovan. Bougainville has always been isolated from Papua by water from what I know. A final question is in regards to Andaman Islanders and other isolated Asian peoples who seem to be hunter-gatherer relics such as the Ainu. Since the Pakistani HGDP populations share a large minor component of ancestry with the Andaman Islanders my assumption is that they should be somewhat deviated toward the Papuans. As the populations are not labeled I do not know if those groups are skewed toward the direction of the Papuans. In the supplements individual outcomes are given for the Han and French, and the Han seem somewhat shifted toward the Bougainville Islanders, though trivially. Additionally, some of the authors of this paper were involved in Reconstructing Indian History, and so I assume had access to Andaman Islander data. I would be curious if they ran some quick checks and decided to stick with the HGDP because there was unlikely to be anything there.

The main body of the paper is tightly and elegantly written. But there is so much more in the supplements. I have read through them at least once, but I can’t say I understand it very well. It is written with the tight economy of a mathematically minded individual, despite the fact that it runs to 90 pages. But much of it alludes to a “D-statistic” which actually goes back to the earlier Neandertal admixture paper, and its supplement. So let’s go back to that, and review the D-statistic at least cursorily. One might not gain a deep knowledge, but even a superficial knowledge of the technical arcana of these sorts of papers are often useful in my experience. To page 130:

To test whether Neandertals share more alleles with some present-day human populations than with others, we compared the Neandertal sequence that we generated to sequence from present-day human samples of diverse ancestry. Specifically, we discovered single nucleotide polymorphisms (SNPs) by comparing exactly two chromosomes from different individuals (H1 and H2). We then assessed whether a test individual (H3, e.g. Neandertal) tended to match either H1 or H2 more often at sites where H3 has the derived allele relative to chimpanzee. Under the null hypothesis that H3 belongs to an outgroup population, it should match H1 and H2 equally often. In contrast, if gene flow has occurred, H3 may match one more than the other.

Here’s a graphical illustration:

The ancestral state is A, which the chimpanzee (not shown as H4) presumably has. B represents the derived state. That means it has changed via mutation from the ancestral state at some point from the last common ancestor with the outgroup. To calculate the D-statistic you are looking at a case where H3 is B and H4 is naturally A. So you have two sets: BABA and ABBA. You are comparing the counts between these two combinations. If H3 is a clean outgroup to the H1H2 clade, D will be ~ 0, as BABA and ABBA counts will approximately be equal. In contrast, if there is gene flow to H1 or H2 from H3, D will deviate from ~0. The Z-score are the standard deviations away from ~0. The table below is from the current paper under consideration. I have highlighted and reformatted:

The D-statistics make sense of what you know verbally. There is some admixture from Neandertals to Eurisicans + Oceanians. Therefore when paired with each other as H1 and H2 they do not deviate as from 0 as much as they do when paired with Africans. There is a deviation away from equal ratios of ABBA and BABA because there is putative gene flow from from H3 to H1 or H2. Notice the Denisovans. Because they’re like Neandertals they produce some elevated deviation from D, though not as much. Interestingly the maximum Z-scores occur when comparing Denisovans, Melanesians, and Africans. Finally, Melanesians and Eurasicans also result in a deviation from 0 when paired with Denisovans in the H3 position.

A quick note from the supplements on ancient population structure. Dienekes does not believe that there was Neandertal admixture necessarily among Eurasicans and Oceanians. From what I can gather he believes that there was population structure within Africa, which is preserved in non-African populations. Rather than exogenous admixture between geographically separated lineages which had only recently met, what one is presumably arguing for here is that there were long term barriers between more closely placed populations in Africa. The authors do not find it parsimonious, though they can not reject it as totally without foundation. Below is a graphical representation of their two models:

So where does this leave us? Yesterday when I said something big was going to drop Ed Brayton expressed some frustration that paleoanthropologists tend to hype stories too much. The reality is everything doesn’t change. The Hobbits, the Darwinius fiasco, and the persistent controversy over Ida, can give anyone fits of human evolution fatigue. But there is a difference here. You don’t need to take their total word for it. At some point you will be able to go to the UCSC genome browser and poke around yourself. Or, you can pull down a 153 MB file with SNPs and indels.

This is a great time to be alive if you’re a hominin natural history nerd. You never know what surprise will greet you when you wake up in the morning. You never know how you’ll have to rearrange your conception of the world. Earlier in the post I mentioned that an instructor once asked a class where I was a student whether scientists should be allowed to talk about the erectine features of Aborigines, if they believed such features existed. You probably won’t be surprised that I said that such things shouldn’t be off limits if they seemed true. Obviously science has political implications. It is idealistic and philosophically consistent to say that it is value-free, but it is also naive. Rather, we need to think hard about how our values relate to the world around us. Or at least some of us need to think hard about that sort of thing.

We shouldn’t take for granted that we all have exactly the same moral intuitions. But on the margin some of our fears are I think overwrought. I know of an individual who admits frankly that they are a “blank slate” maximalist because they don’t know how they could sleep or live if many traits had some hereditary component. Similarly, I have met many conservative Christians and Muslims who admit that they would rape, murder and steal if they didn’t believe in God. In other words, if God doesn’t exist they would become psychopaths, because “why not.” This is ludicrous. God doesn’t exist, and they aren’t psychopaths. They may believe that they aren’t sodomizing their sister because the Lord God declared from On High believes that such behavior is forbidden, but I think that’s ridiculous on the face of it (on the margin there may be some effect of belief in God on behavior by the way, but that’s not what I’m getting at here obviously). Everything may be possible, but everything is not palatable. As for the possibility that humans may differ substantially from individual to individual and group to group, if you acknowledge this one day will you then as a matter of course raise in your arms in salute? If so, it is true that humans differ profoundly in matters of moral sense, because I could not comprehend such behavior.

So Papuans, and likely Aborigines, are likely ~7.5% non-neo-African. Does that matter? Do they bleed today where yesterday they did not? In deep matters of substance nothing is different from this moment than before. Let me quote John Hawks:

Our common ancestry as humans goes back to the Early and Middle Pleistocene. The (now multiple) Neandertal genomes and the Denisova genome share genes with some people and not others because of this common ancestry.

In addition, some living people carry even more genes from Neandertals because they have an appreciable fraction of Neandertal ancestry. That makes it nonsensical to talk about “Neandertals and the ancestors of modern humans”. Neandertals are among the ancestors of modern humans.

Just so with Denisova. It’s nonsensical to talk about a three-way split between Neandertals, Denisova and modern humans. We can talk about a population model with a clade separating an ancestral Neandertal-Denisova population from contemporary Africans.

I have to remind myself again and again when I talk to people about these issues that “modern human ancestors” is not a group that excludes these Pleistocene people.

Once we put ourselves into the mode where we are referring to a population model, it is important to recognize the limitations of those models. For example, we cannot presently exclude many kinds of gene flow among these Pleistocene populations. We can understand some limits to the level of gene flow — these populations were highly structured, it wasn’t Pleistocene panmixia. But it is premature to talk about isolation without recognizing the limits of our ability to test these population models.

The difficulty with terminology tells us something very important. A large-scale reorganization of the science of human origins is upon us. The terms we are used to using will, many of them, become obsolete. Some now-obscure terms will become very important.

What we know to be good and true is still good and true. It is a small soul who is so moved by matters of terminology, we should be cautious of allowing that to happen to ourselves. I think now to the fact that both the Romans and Muslims abhorred the idea of the king. The Romans overthrew their monarchy, established a republic, and replaced it with a despotism which was a monarchy in all but name. The Muslims had caliphs, vice-reagents of God, and sultans and emirs, who were vice-reagents of the caliphs. Despite the glory which is given over to their God the Muslim despotisms were things of men. Domination of the many by one is a matter of substance, not style. Human dignity should not be contingent on details of ancestry. Isn’t that obvious? I thought that was what the 20th century was to some extent all about.

Back to the science. I began with a long historical sketch, viewed through my own personal lens, because probabilities are always filtered through a glass of accreted priors. I was not as shocked by many at the idea of intogression and admixture because Greg Cochran, Henry Harpending, and John Hawks had already predisposed me to think about the plausibility of such phenomena. Additionally, I have always had an interest in conservation genetics, as well as modeling cultural evolution. Such lateral flows are not unknown in those domains. When I first discussed the Neandertal admixture results with Oren Harman last spring he reminded me that one should be cautious of such things; many splashy science stories often don’t pan out. And yet with all due respect to Oren, in this case we do need to observe that there has been a veritable mob of scholars pouring over these data. Additionally, this is something old, not something new.

These results will not remain isolated findings with only parochial relevance. I believe these two papers will probably shift the equilibrium orthodoxy in a new direction. Old models and genetic studies will be seen in a new light. Anomalies unconsidered will get a second look. In The New York Times Stanford geneticist Carlos Bustamante seemed to indicate to Carl Zimmer that the hunt was on. Perhaps the human genome is more of a mosaic than we thought?

Finally, one wonders how this was missed. 7.5% is not trivial. And yet a generation of mtDNA and NRY studies have seemingly missed this. I presume that the archaic admixture didn’t show up in STRUCTURE because it’s a stabilized part of the genetic background of Eurasicans and Oceanians. It reminds of us the limitations of interpretation. We know what we know contingent on what we already know. Since we know more, a different set of inferences may now be generated. Though with due humility. Not quite time yet for the hardening of a new orthodoxy.

Personal note: Merry Christmas! Obviously it is time for me to take a break. Best wishes, and let’s make 2011 more informative and data rich. Hopefully we won’t have to wait too long for Otzi’s genome.

Citation: Reich, David, Green, Richard E., Kircher, Martin, Krause, Johannes, Patterson, Nick, Durand, Eric Y., Viola, Bence, Briggs, Adrian W., Stenzel, Udo, Johnson, Philip L. F., Maricic, Tomislav, Good, Jeffrey M., Marques-Bonet, Tomas, Alkan, Can, Fu, Qiaomei, Mallick, Swapan, Li, Heng, Meyer, Matthias, Eichler, Evan E., Stoneking, Mark, Richards, Michael, Talamo, Sahra, Shunkov, Michael V., Derevianko, Anatoli P., Hublin, Jean-Jacques, Kelso, Janet, Slatkin, Montgomery, & Paabo, Svante (2010). Genetic history of an archaic hominin group from Denisova Cave in Siberia Nature : 10.1038/nature09710

November 24, 2010

The inevitable social brain

ResearchBlogging.orgOne of the most persistent debates about the process of evolution is whether it exhibits directionality or inevitability. This is not limited to a biological context; Marxist thinkers long promoted a model of long-term social determinism whereby human groups progressed through a sequence of modes of production. Such an assumption is not limited to Marxists. William H. McNeill observes the trend toward greater complexity and robusticity of civilization in The Human Web, while Ray Huang documents the same on a smaller scale in China: A Macrohistory. A superficial familiarity with the dynastic cycles which recurred over the history of Imperial China immediately yields the observation that the interregnums between distinct Mandates of Heaven became progressively less chaotic and lengthy. But set against this larger trend are the small cycles of rise and fall and rise. Consider the complexity and economies of scale of the late Roman Empire, whose crash in material terms is copiously documented in The Fall of Rome: And the End of Civilization. It is arguable that it took nearly eight centuries for European civilization to match the vigor and sophistication of the Roman Empire after its collapse as a unitary entity in the 5th century (though some claim that Europeans did not match Roman civilization until the early modern period, after the Renaissance).

It is natural and unsurprising that the same sort of disputes which have plagued the scholarship of human history are also endemic to a historical science like evolutionary biology. Stephen Jay Gould famously asserted that evolutionary outcomes are highly contingent. Richard Dawkins disagrees. Here is a passage from The Ancestor’s Tale:

…I have long wondered whether the hectoring orthodoxy of contingency might have gone too far. My review of Gould’s Full House (reprinted in A Devil’s Chaplain) defended the popular notion of progress in evolution: not progress towards humanity – Darwin forend! – but progress in directions that are at least predictable enough to justify the word. As I shall argue in a moment, the cumulative build-up of compelx adaptations like eyes strongly suggest a version of progress – especially when coupled in imagination with of the wonderful products of convergent evolution.

Credit: Luke Jostins
Credit: Luke Jostins

One of those wonderful products is the large and complex brains of animals. Large brains are found in a disparate range of taxa. Among the vertebrates both mammals and birds have relatively large brains. Among the invertebrates the octopus, squid and cuttlefish are rather brainy. The figure to the right is from Luke Jostins, and illustrates the loess curve of best fit with a scatter plot of brain size by time for a large number of fossils. The data set is constrained to hominins, humans and their ancestors. As you can see there is a general trend toward increase cranial capacities across all the human populations. Neandertals famously were large-brained, but they exhibited the same secular increase in cranial capacity as African Homo. On the scale of Pleistocene Homo and their brains the idea of the supreme importance of contingency seems ludicrous. Some common factor was driving the encephalization of humans and their near relations over the past two million years. This strikes me as very strange, as the brain is metabolically expensive, and there are plenty of species with barely a brain which are highly successful. H. floresiensis may be a human instance of this truism.

But what about the larger macroevolutionary pattern? Is there a trend toward larger brain sizes in general, of which primates, and humans in particular, are just the most extreme manifestation? Some natural historians have argued that there is such a trend. But, there is a question as to whether increased brain size is simply a function of allometry, the pattern where different body parts and organs tend to correlate together in size, but also shift in ratio with scale. The nature of physics means that very large organisms have to be more robust because their mass increases far faster than their surface area. By taking the aggregate relationship between body size and brain size, and examining the species which deviate above or below the trend line, one can generate an encephalization quotient. Humans, for example, have a brain which is inordinately large for our body size.

And yet there are immediate problems looking at relationships between body and brain size, and inferring expectations. Different species and taxa are not interchangeable in very fundamental ways, and so a summary statistic or trend may obscure many fine-grained details. A new paper in PNAS focuses specifically on various mammalian taxa, corrects for phylogenetics, and also relates encephalization quotient by taxa to the proportion of social animals within each taxon. Encephalization is not a universal macroevolutionary phenomenon in mammals but is associated with sociality:

Evolutionary encephalization, or increasing brain size relative to body size, is assumed to be a general phenomenon in mammals. However, despite extensive evidence for variation in both absolute and relative brain size in extant species, there have been no explicit tests of patterns of brain size change over evolutionary time. Instead, allometric relationships between brain size and body size have been used as a proxy for evolutionary change, despite the validity of this approach being widely questioned. Here we relate brain size to appearance time for 511 fossil and extant mammalian species to test for temporal changes in relative brain size over time. We show that there is wide variation across groups in encephalization slopes across groups and that encephalization is not universal in mammals. We also find that temporal changes in brain size are not associated with allometric relationships between brain and body size. Furthermore, encephalization trends are associated with sociality in extant species. These findings test a major underlying assumption about the pattern and process of mammalian brain evolution and highlight the role sociality may play in driving the evolution of large brains.

A key point is that the authors introduce time as an independent variable, so they are assessing encephalization over the history of the taxon. This is clearly relevant for humans, but may be so for other mammalian lineages. The table and figures below show the encephalization slope generated by using time and body size as the predictors and brain size as the dependent variable. A positive slope means that brain size is increasing over time.

Two major points:

- Note that the slope is sensitive to the level of taxon one is examining. A closer focus tends to show more variance between taxa. So, for example, humans distort the value for primates in general. Bracketing out anthropoids paints a more extreme picture of encephalization, a higher slope. In contrast, the lemurs and their relatives exhibit less encephalization over time.

- The correlation between proportion of species which exhibit sociality and encephalization of the taxon is strong. From the text:

Encephalization slopes were correlated with both the proportion of species with stable groups (order R = 0.92, P = 0.005, n = 6; suborder R = 0.767, P = 0.008, n = 9; Fig. 2 A and B) and the proportion in either facultative or stable social groups (order R = 0.804, P = 0.027, n = 6; suborder R = 0.63, P = 0.04, n = 9).

The last figure makes it is clear that the correlations are high, so the specific values should not be surprising. Don’t believe these specific figures too much, how one arranges the data set or categorizes may have a large effect on the p-value. But the overall relationship seems robust.

A highly encephalized “alien”

What to think of all of this? If you don’t know, one of the authors of the paper, Robin Dunbar, has been arguing for the prime importance of social structure in driving brain evolution among humans for nearly twenty years. The relationship is laid out in his book Grooming, Gossip, and the Evolution of Language. Robin Dunbar is also the originator of the eponymous Dunbar’s number, which argues that real human social groups bound together by interpersonal familiarity have an upper limit of 150-200. He argues that this number arises because of the computational limits of our “wetware,” our neocortex. Those limits presumably being a function of biophysical constraints.

One interesting fact though is that the median cranial capacity of our species seems to have peaked around one hundred thousand years ago. The average human today has a smaller brain than the average human alive during the Last Glacial Maximum! (see this old post from Panda’s Thumb, it’s evident in the charts) This may be simply due to smaller body sizes in general after the Ice Age. Or, it may be due to the possibility that social changes with the rise of agriculture required less brain power.

Ultimately if Dunbar and his colleagues are correct, if social structure is the most powerful variate in explaining differences in brain size when controlling for phylogenetics and body size, then in some ways it is surprising to me. After all, it does not seem that ants have particularly large brains, despite being extremely social and highly successful. Clearly the hymenoptera and other social insects operate on different principles from mammals. Instead of
developing “hive minds,” it seems as if in mammals greater social structure entails greater cognitive structure.

Citation: Susanne Shultz, & Robin Dunbar (2010). Encephalization is not a universal macroevolutionary phenomenon in mammals but is associated with sociality PNAS : 10.1073/pnas.1005246107

September 22, 2010

Toward human phylogenetic intuitions

Filed under: Genetics,Genomics,Human Population Structure,phylogenetics — Razib Khan @ 1:37 am


100 years ago a science based physical anthropology offered up very little as to a systematics of mankind beyond what you could intuit from visual assessments of phenotypic similarity alone. Instead, there were fantastical taxonomies which had little basis in the true pattern of variation and more in the nationalistic debates of that period. The Nordic, Mediterranean, and Alpine trichotomy of the European peoples had only marginally more concrete reality than the division between the Vanyar, Noldor, and Teleri.

We don’t live in such a fantastic age. Much of the mystery, and so potential for mischief, is gone. The “post-genomic” era means that old questions only vaguely perceived in the past are now well resolved. Quite often readers will ask a question as to the phylogenetic relationship between population A & B. If I don’t know off the top of my head, which is the norm, I’ll go to the search engine and look up what I’ve written on the topic. This has started to become tedious, in part because Wordpress’ search engine leaves something to desired. So I have some papers bookmarked for immediate reference. They’re of wide scope (i.e., they don’t focus on just one population such as the Jews) and draw from a large number of markers to get a good picture of total genome relatedness. The focus within these papers tends to be genetic distances and relationships, not other topics of great interest such as natural selection. Also, I’ve tried to find links accessible to people without institutional access (for the Science link free registration will do it). If you can think of other papers, please leave the link in the comments.

Worldwide human relationships inferred from genome-wide patterns of variation

Genome-wide Insights into the Patterns and Determinants of Fine-Scale Population Structure in Humans

Toward a more uniform sampling of human genetic diversity: A survey of worldwide populations by high-density genotyping

The Population Reference Sample, POPRES: A Resource for Population, Disease, and Pharmacological Genetics Research

The Genetic Structure and History of Africans and African Americans

Genes mirror geography within Europe

Genomic Dissection of Population Substructure of Han Chinese and Its Implication in Association Studies

Reconstructing Indian population history

Whole-Genome Genetic Diversity in a Sample of Australians with Deep Aboriginal Ancestry

Genetic Variation and Population Structure in Native Americans

September 10, 2010

Leaping to a bigger brain

800px-RedRooYears ago an evolutionary biologist mentioned to me almost offhand that with the emergence of genomics and the necessity to master computational techniques a lot of the labor hours which may have gone into a more thorough understanding of specific organisms had gone by the wayside. He believed that his Ph.D. advisor was going to take a lot of knowledge with him when he retired because there was just no time to devote to discussing details of specific organismic life history, anatomy, and behavior. I obviously think that the sacrifice has been worth it, the new methods are powerful and answer long standing questions (or hold promise to do so), but something has no doubt been lost. Biological variation is such that a gestalt “big picture” sense of the lay of the land is useful. Much of biology is a historical science, and like history the details are of the essence. But unlike history biology is a natural science, and amenable to experimentation and observation, as well as laced with a more thorough formalization (yes, I am aware of cliometrics). The mileage one gets out of theory in biology is far greater than in history, as evidenced by the high prestige of an evolutionary framework, and the obscurity of cliodynamics (and the relative marginal reputation of Arnold Toynbee).

But evolution purely as logic often fails. The old debate between the balance & classical schools in evolutionary genetics was upended by empirical findings in molecular evolution in the 1960s, which subsequently stimulated neutral theory. Natural science has to extend itself through a long-term dance between system building and empirical verification or falsification. The seeds of new systems don’t come from a vacuum, rather, the prior set of observations and experiments lay the groundwork and serve as points of embarkation.

ResearchBlogging.orgThe combination of biology’s variation and its reliance on theories, heuristics, and rules-of-thumb (e.g., 19th century biology’s love affair with “laws”), often leads to perplexing surprises when a more systematic or deeper read of the data flies in the face of expectations. So it is with a new paper in PNAS which upends some specific relationships between mammalian characteristics and encephalization, as well as some more general prejudices. Brain size, life history, and metabolism at the marsupial/placental dichotomy:

The evolution of mammalian brain size is directly linked with the evolution of the brain’s unique structure and performance. Both maternal life history investment traits and basal metabolic rate (BMR) correlate with relative brain size, but current hypotheses regarding the details of these relationships are based largely on placental mammals. Using encephalization quotients, partial correlation analyses, and bivariate regressions relating brain size to maternal investment times and BMR, we provide a direct quantitative comparison of brain size evolution in marsupials and placentals, whose reproduction and metabolism differ extensively. Our results show that the misconception that marsupials are systematically smaller-brained than placentals is driven by the inclusion of one large-brained placental clade, Primates. Marsupial and placental brain size partial correlations differ in that marsupials lack a partial correlation of BMR with brain size. This contradicts hypotheses stating that the maintenance of relatively larger brains requires higher BMRs. We suggest that a positive BMR–brain size correlation is a placental trait related to the intimate physiological contact between mother and offspring during gestation. Marsupials instead achieve brain sizes comparable to placentals through extended lactation. Comparison with avian brain evolution suggests that placental brain size should be constrained due to placentals’ relative precociality, as has been hypothesized for precocial bird hatchlings. We propose that placentals circumvent this constraint because of their focus on gestation, as opposed to the marsupial emphasis on lactation. Marsupials represent a less constrained condition, demonstrating that hypotheses regarding placental brain size evolution cannot be generalized to all mammals.

The prejudice, which they mention in the text, is that marsupials are relatively small-brained vis-a-vis placentals. Though modern biology long ago rejected an explicit Great Chain of Being, I still think it is very hard to avoid the implicit working model that marsupials are somehow fundamentally inferior to placentals, and their stronghold in Australia and Melanesia is a manner of historical contingency (I believe this is an outgrowth of natural anthropocentrism). Additionally the authors claim that the fixation on kangaroos and their relatives as the exemplary marsupials has also given an improper impression of the clade’s encephalization. Similarly, the inclusion of primates distorts the overall perception of encephalization among placentals. The first figure makes clear as a descriptive matter the reason for our misimpression. I’ve reedited and labeled for clarity.


Phylogenetic generalizations are naturally sensitive to the clades you are evaluating across. Prior to reading this paper my own understanding was that placentals were invariably larger-brained for their body size than marsupials. This is true still. But to a great extent this is simply an outcome of the placement within the placental clade of primates, a group which is atypical for mammals as a whole, placental or marsupial.

Next they looked at the log-transformed values of three traits against body mass in grams. I’ve rotated the figure, but you see panel A, B, and C, brain weight, BMR, and gestation + weaning age.


Biological organisms are constrained to some extent by physical parameters, so all mammals follow roughly the same scaling trends. But, there are differences between placentals, marsupials, and monotremes. Interestingly smaller marsupials, those below 43 grams, are more encephalized than placentals of the same size. Placentals tend to have somewhat higher basal metabolic rates, BMR, than marsupials, and especially monotremes. Finally, the last figure shows a looser trend line with a larger residual. I suspect it’s because it is more about life history than a concretely physical parameter. In any case, here marsupials are above the placental trend line, in large part because marsupial young have a long phase of dependence on the mother after birth. In contrast placentals invest much more time and energy on gestation. This period is especially important for the question of encephalization, since so much of brain growth and development occurs during this early phase.

The third figure and the first table show the partial correlations between the traits of interest, and the regression slopes for the various traits in terms of their prediction of encephalization. I’ve reedited the figure and table a bit, including only the phylogenetically corrected regression. Filled arrows indicate positive correlations, and open arrows negative ones. The partial correlations are separate also for marsupials and placentals:


A quick scan of the regression table shows that marsupials and placentals seem to show very different relationships between the predictors and encephalization. BMR is predictive of encephalization in placentals, and explains about ~10% of the variation (the r-squared). There is no statistically significant relationship in marsupials. Rather, ~20% of the variation in encephalization in marsupials is predicted by weaning age. Knowing what we know about marsupials, that they have relatively short gestations and long periods of dependence on maternal provisioning through lactation, this seems unsurprising. Weaning age is important in placentals, but less so. Rather, some of the “slack” seems to be taken up by gestation amongst them. Litter size is a strong negative predictor of encephalization. This seems to follow our intuition. There’s a trade off between quality and quantity of offspring.

It isn’t just placentals and marsupials that the authors contrast in this paper. They bring up another vertebrate lineage which has evolved toward relatively large brains and high basal metabolic rates: birds. Like marsupials the correlates of greater encehpalization among birds seems to be rather different from placentals. This is fundamentally a story of different strokes. Next I’d like to see cephalopods added to the equation! Rather than going into the details any further, let me jump to the conclusion:

Our results confirm several hypotheses of mammalian brain size evolution—in particular, the prediction of the maternal energy hypothesis that large-brained mammals with lower BMRs should have extended maternal investment times. In addition, our inclusion of marsupials provides further insight into the patterns of mammalian brain size evolution by showing that placental brain size evolution represents a unique case among mammals, connected with the placental reproductive emphasis on gestation. Based on this, several avenues for further research arise. If BMRs exceed brain maintenance rates in extant mammals, investigation of brain size in mammalian ancestors will provide clues as to when (and perhaps how often) the minimum BMR to allow a mammalian-sized brain evolved. Due to the close interaction between reproduction related brain size correlates and brain ontogeny, an improved understanding of brain growth and structural development patterns in species with different reproductive strategies emerges as another important area of future research. Our results emphasize that factors influencing the evolution of brain size are complex and emerge from fields that are traditionally researched separately, such as physiology, developmental biology, zoology, and paleontology. The integration of such interdisciplinary research represents the most appropriate avenue for providing a comprehensive evolutionary background for neurobiological research.

What if some of what you knew was wrong? To me if this paper checks out that’s going to be a take home for me. I simply assumed that placentals were highly encephalized in relation to marsupials. That seems not to be the case. Additionally, it seems that the trait level correlations of encephalization vary by phylogenetic clade, so that there are different ways to ascend the peaks of the big-brained. I take a deep interest for a very human reason, this figure generated by Luke Jostins:
linearThe figure shows the encephalization over time of various human lineages. In particular, it shows that disparate branches of the human family tree were all simultaneously increasing their cranial capacities until about ~200,000 years ago. Why? To understand, or at least generate some plausible explanations, we need to get a better grasp of encephalization across mammals, and further out across the animal kingdom. Recall that new systems emerge from previous patterns of observations. Primates are already the outliers among placental mammals, and we are the outliers among primates. Something very strange has been going on across our lineage for the past two million years, and we don’t quite know yet. Don’t tell Robert Wright, but I’m beginning to wonder if humanity was inevitable. At least after a certain point.

Citation: Weisbecker V, & Goswami A (2010). Brain size, life history, and metabolism at the marsupial/placental dichotomy. Proceedings of the National Academy of Sciences of the United States of America PMID: 20823252

Image Credit: Wikimedia

May 20, 2010

Chickens are like people

Filed under: Genetics,Genomics,phylogenetics — Razib Khan @ 3:17 am

In that their demographic history is complicated. The Origin and Genetic Variation of Domestic Chickens with Special Reference to Junglefowls Gallus g. gallus and G. varius:

… domestic chickens diverged from red junglefowl 58,000±16,000 years ago, well before the archeological dating of domestication, and that their common ancestor in turn diverged from green junglefowl 3.6 million years ago. Several shared haplotypes nonetheless found between green junglefowl and chickens are attributed to recent unidirectional introgression of chickens into green junglefowl. Shared haplotypes are more frequently found between red junglefowl and chickens, which are attributed to both introgression and ancestral polymorphisms. Within each chicken breed, there is an excess of homozygosity, but there is no significant reduction in the nucleotide diversity. Phenotypic modifications of chicken breeds as a result of artificial selection appear to stem from ancestral polymorphisms at a limited number of genetic loci.

I wonder if domesticates in particular exhibit these more complex reticulated patterns in their phylogenies because they spread along human trade routes.

March 30, 2010

The ways of the forefathers & foremothers

Filed under: Anthroplogy,Cultural Evolution,Culture,phylogenetics — Razib Khan @ 6:17 pm

Fascinating post by Bayes, Phylogenetics, cultural evolution and horizontal transmission:

For some time now, evolutionary biologists have used phylogenetics. It is a well-established, powerful set of tools that allow us to test evolutionary hypotheses. More recently, however, these methods are being imported to analyse linguistic and cultural phenomena. For instance, the use of phylogenetics has led to observations that languages evolve in punctuational bursts, explored the role of population movements, and investigated the descent of Acheulean handaxes. I’ve followed the developments in linguistics with particular interest; after all, tracing the ephemeral nature of language is a daunting task. The first obvious road block is that prior to the invention of writing, the uptake of which is limited in geography and history, language leaves no archaeological record for linguists to examine. One particular note I’d like to make is that when Charles Darwin first formulated his theory of natural selection, he took inspiration from linguistic family trees as the basis for his sketch on the evolutionary tree of life. So it seems rather appropriate that phylogenetic approaches are now being used to inform our knowledge regarding linguistic evolution.

Like many other attempts applying evolutionary thinking in culture, phylogenetic approaches are, at times, met with contempt. This stems from assertions that cultural evolution and biological evolution differ greatly in regards to the relative importance of horizontal transmission….

I guess the general points to take away from this post are: 1) Do not necessarily assume horizontal transmission is dominant in shaping culture; and, 2) Even with certain levels of reticulation, it does not necessarily invalidate a phylogenetic approach in investigating cultural and linguistic evolution.

I think the point that horizontal transmission may be less important relative to vertical transmission than we’d previously thought in regards to the spread and diffusion of cultures may explain some of the recent findings from DNA extractions which suggest that hunter-gatherers were replaced in Europe by farmers. The standard model before the recent wave of extractions was that farming spread through cultural diffusion, with a minority view championed by L. L. Cavalli-Sforza of “demic diffusion” whereby demographic growth from the point of origination spread a culture, though the initial distinctive genetic signal became progressively weaker through dilution via admixture. But if cultural practices such as agriculture were much more vertically transmitted, from parent to child, rather than horizontally across societies, the genetic pattern of replacement becomes more comprehensible.

Of course, the main caveat is that intermarriage has been very common between neighboring groups. The rape of the Sabine women may reflect a common practice on the part of migratory males; the Greek colonization of the western Mediterranean was almost all male, so the subsequent generations were biologically the products of Greek men and native women (though culturally they were fully Greek, as evidenced by the term “Magna Graecia” to refer to Sicily and southern Italy). It is not atypical for vertical transmission of culture to occur from one parent, and in particular one sex. More recently the descendants of the pairings of Iberian men and indigenous women in Latin America tend to speak Spanish and avow the Christian faith. Though aspects of local identity, such as cuisine and clothing, may retain an indigenous stamp it is no coincidence that these populations are labelled “Latin American” despite their mixed genetic provenance.

Note: In the United States children have traditionally been more often raised in the denomination of their mother than father, so there isn’t always a male-bias in vertical transmission when the parents are not concordant for a cultural trait.

March 27, 2010

The Mysterious Other

Last week Nature published a paper which may have found a new ‘branch’ of the hominin evolutionary bush which may have been coexistent which modern humans and Neandertals. I recommend The Atavism, Carl and John Hawks on this story. Interesting times.

Powered by WordPress