The Bushmen tell us a lot about human evolution because they are humans who have evolved

By Razib Khan | September 21, 2012 12:32 am

When it comes to the human genetics of the Khoe-San there’s a little that’s stale and unoriginal for me in terms of presentation. The elements are always composed the same. The Bushmen are the “most ancient” humans, who can tell us something about “our past,” about “our evolution.” Tried & tested banalities just bubble forth unbidden. I have no idea why. There’s a new paper in Science on the genetics of the Khoe-San, which includes Bushmen, which brought to mind this issue for me because of the outrageous nature of the press releases.

The title of the paper itself is a testament to vanilla, Genomic Variation in Seven Khoe-San Groups Reveals Adaptation and Complex African History. This is absolutely not surprising. Are you shocked that the Khoe-San have adaptations? Or that African history is complex? The wonder of it all! This paper actually revisits much of the same ground as Pickrell et al.’s originally titled The genetic prehistory of southern Africa. Before Dr. Pickrell executes throw-down on me on Twitter let me concede that I have no creative ideas to offer in terms of an alternative title. Rather, I have an idea: perhaps in the future scientists could explore the evolutionary genetic basis for steatopygia? The trait is not limited just to Khoe-San, my distant cousins the Andaman Islanders also exhibit it. Perhaps this is the ancestral state of the human lineage? This is a situation where the titles just write themselves!


I bring up this point because from what I can tell the main area where the Science paper adds anything to the one posted on arXiv is that there is some exploration of adaptation and natural selection. In particular, the authors find that not only are pastoralist Khoi groups admixed with Bantu, but they seem to have lactase persistence alleles shared with East African populations! And, there is a pigmentation locus which seems under strong selection in them, introgressed from the Bantu. I am not entirely familiar with one of their tests for selection, which proposes to explore very deep time adaptive events (on human evolutionary scales), so I will comment no further. But I am intrigued that there seems to be a worldwide tendency for their to be selection on pigmentation genes.

Though the paper itself is gated, you can really read most of the good stuff in the supplements. I believe that is free to non-subscribers (you may have to register for free, I don’t know). I get a feeling that the group that published in Science took a “kitchen sink” approach to rummaging through their data set. Pickrell et al.’s preprint (now accepted to Nature Communications) seems positively spare at an economically written 63 pages. I would concede that the prose in the gated text is more polished, but that’s only a small proportion set next to the sometimes rambling supplement. An interesting window into the possible outcomes of different forms of publication.

My primary scientific concern about Genomic Variation in Seven Khoe-San Groups Reveals Adaptation and Complex African History is what I see as reliance on assumptions in their phylogenetic methods. I’m not sure that their values hold, though in its broad outlines their natural history of the Khoe-San seem to support what has previous been inferred from mtDNA lineages, and the whole genome analysis of the 2009 Bushmen paper (interestingly, the first author of the current Science paper sent in a letter objecting to the terminology used in that 2009 paper, and it seems she attempted to be as punctilious as possible in this area by making sure there was a robust ethnography). Let me quote:

A third factor that, in contrast to the other two, lead to an overestimation of the population divergence time of the Khoe-San to other groups is admixture from more basal (’archaic’) human population lineages (120). An important direction of future research is thus to investigate the extent of ancestral structure in these populations, but assuming that such contributions are relatively minor (120), we suggest that the majority of the ancestry present Khoe-San groups today is from a population that diverged from a population lineage leading to most of the ancestry of all other modern humans at least 100,000 years ago

Why assume that? In fact the ScienceDaily summation blares out “Khoe-San Peoples Diverged Before ‘Out-Of-Africa’ Migration of Modern Humans”. First, there was a lot of evidence pointing in this direction before this paper. Second, that inference rests upon that assumption, at least to some extent. Overall I wasn’t totally convinced that the authors had done a thorough enough job accounting for…well, the complexity of African genetic history in regards to admixture.  Trees have limits at this remove in time.

This is an easy gripe to have. I understand the constraints to comptutation, as well as the labor hours necessary, and the fact that this is a scientific paper, not a magnum opus. But I still have to wonder if perhaps I would be less irritated on this measure if the authors hadn’t simply hoovered up every cutting-edge statistical genetic method of inference and presentation and repackaged it with their data, rather than focusing on and refining their phylogenetic techniques. You don’t need to be an academic at a research institution to run ADMIXTURE or EIGENSOFT. How many PCA plots of African populations do you really need?

This paper, and The genetic prehistory of southern Africa, make me come back to my thoughts about the origin of ‘modern humans,’ as well as the Khoe-San. If there were Khoe-San evolutionary geneticists around I wonder if they’d be talking about how non-Khoe-San were an interesting side branch of the Khoe-San, and offered a window into the evolutionary possibilities of our species. We tend to imagine our species as having arisen in a singular manner, de novo and sui generis, fully formed from the head of Zeus. If the Khoe-San diverged very early on after this miraculous event, then a particular interest in them makes sense. But today with all the talk of admixture and numerous archaic hominins in the family tree of our own present lineage, I think perhaps we need to just let go of this idea, and let the Khoe-San just be, rather than be a window into some seminal evolutionary event. They’re our distance cousins, but perhaps our common grandparents just weren’t that special.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    two points

    – wanted to know more about the mbuti-khoe-san comparisons. i’m intrigued, but unconvinced. high uncertainty.

    – the great divergence among the bushmen intrigue me. have these populations been local to this region for 20-40 thousand years???

  • http://abugblog.blogspot.com Africa Gomez

    Brilliant post “If there were Khoe-San evolutionary geneticists around I wonder if they’d be talking about how non-Khoe-San were an interesting side branch of the Khoe-San, and offered a window into the evolutionary possibilities of our species” Loved that! a pity is longer than 140 characters. There are two branches and non Khoe-San are not a subset of Khoe-San diversity. Also hate the press releases about Khoe-San being descendants of oldest humans. So silly.

  • Grey

    “We tend to imagine our species as having arisen in a singular manner”

    Given all the distinct environments in Africa i’d have thought there might be lots of early branches each adapted slightly differently for their particular environment and the initially most successful (in terms of surface area or numbers) and the eventual winner aren’t neccessarily the same. For example i wonder how much the Bantu expansion may have over-written what was there before.

  • Eurologist

    ” the great divergence among the bushmen intrigue me. have these populations been local to this region for 20-40 thousand years???”

    Well, local to some region. Africa is a big place, and there is no reason to believe that populations today lived there 40,000 ya – we don’t make that assumption anywhere else on the planet. Also, admixtures with ancient populations could have been (and probably were) highly local, and movement in paleolithic times was likely highly restricted, with the additional problem that people originating in contact regions during good climatic times would have a high tendency to become genetically irrelevant during bad times (when the non-admixed core populations survive).

  • Paul

    I like that research topic and I cannot lie.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #4, yes, that’s what i’m getting at.

  • https://sites.google.com/site/josephpickrell/ J Pickrell

    Nice post. I should make it very clear that our group had nothing to do with the press release by the other group.

    If there were Khoe-San evolutionary geneticists around I wonder if they’d be talking about how non-Khoe-San were an interesting side branch of the Khoe-San, and offered a window into the evolutionary possibilities of our species.

    Excellent point. I wonder if this comes down to the well-known issues people have reading phylogenetic trees. We have a figure (see this post) where we have the (many) Khoisan groups displayed below a (small) set of non-Khoisan groups. I’ve already had people ask me how it is that we infer non-Khoisan groups to have split off earlier in human history! I guess in this visualization the rest of humanity is a special side-branch of the Khoisan.

  • https://sites.google.com/site/josephpickrell/ J Pickrell

    the great divergence among the bushmen intrigue me. have these populations been local to this region for 20-40 thousand years???

    There’s a hypothesis that agriculturalist migrations to southern and eastern Africa caused major population movements among hunter-gatherer groups (which must be true to some extent). Perhaps the different Khoisan groups were geographically quite distant until the last few thousands years, and the Kalahari is a “refuge” of sorts.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #8, that was my thought basically. and the similarities might be due to subsequent gene (and meme) flow?

  • https://sites.google.com/site/josephpickrell/ J Pickrell

    According to the linguists I know, the languages of the SE (Tuu) and NW (Ju) Khoisan populations have no known relationship, so as far as they can tell there aren’t even that many similarities.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    ok then, i wonder, are the phenotypic similarities between distant bushmen groups then simply just perceived by the rest of humanity? or perhaps they are some sort of local adaptation? (you know at least one of the phenotypic adaptations i’m getting at….)

  • toto

    But I am intrigued that there seems to be a worldwide tendency for their to be selection on pigmentation genes.

    Actually what I would find intriguing is that the supposedly autochtonous Khoi would find benefit in a pigmentation gene from supposedly Tropical Bantus?

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #12, yep. i wonder if the pigmentation is not the target of selection at all? the mutation results in more burning!

  • https://sites.google.com/site/josephpickrell/ J Pickrell

    I’m actually not aware of any physical anthropology work comparing the NW and SE Kalahari Khoisan groups, and personally have never been to this part of the world.

    There are reports that epicanthic folds are common in the Sandawe (east Africa) and the Khoisan (unclear which Khoisan groups). I assume it is non-trivial to organize a large GWAS in this part of the world, but if someone found the locus (loci?) in east Asians, it’s possible that could be an interesting trait to study in terms of local adaptation.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    re: epicanthic fold. i’m intrigued because i think the assumption is that it is a derived trait in northeast asians (drops in frequency away from korea; you can simply correlate with decrease in demand for eyelid surgeries ;-) what if not? if it is derived it seems likely that it will be different alleles…i think it segregates at a large effect QTL in asians. it seems to vary in a dichotomous fashion in families.

  • Dwight E. Howell

    I would like to see more research done on Steatopygia. Clearly an adaption to food shortages in warm climates unless male sexual preference drove this well past the point of adaption to physical circumstances.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #16, a lot of ppl would like to see more research on that i suspect.

  • Insightful

    Aye, Pickrell and Razib – the epicanthic eyefold is also found among the Nuba people in Sudan. Take a look:

    http://www3.picturepush.com/photo/a/9928096/1024/Anonymous/Nuba-Woman.bmp

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #18, interesting. i had no idea. one old explanation is that they were an adaptation to cold or something. seems not.

  • Sandgroper

    The story I heard was so they can get around in sand storms in the Gobi Desert or something. Obviously equally nonsensical. I’d be surprised if it is an adaptive trait.

  • https://sites.google.com/site/josephpickrell/ J Pickrell

    #18, I’m with Razib, didn’t know that, very interesting. Someone should map it in one of these populations!

  • silylene

    I was also noticing that the Khoisan have relatively flat faces (low mandibular prognathism), which is more similar to Eurasians and some east african groups, but different than sub-saharan west african groups who tend to have a more prominent prognathism. This makes me wonder what is the origin of prognathism in sub-saharan west african groups ?

  • ogunsiron

    I have a female cousin who’s pretty much purely black but yet has eyes like a north east-asian. It’s not common by any means, but it doesn’t surprise me when a purely black person has eyes like that.

  • Sandgroper

    No, not at all, I have seen it in a lot of Africans. The problem is in remembering which ones. Do you know which part of Africa she derived from?

    If it is not an adaptive trait, it means it has been transmitted somehow fron African groups to north-east Asians, but not to others, assuming it arises the same way, so it could be important information. My wife is ancestral northeast Chinese and she has fairly pronounced epicanthic folds.

  • silylene

    Epicanthic folds are also a symptom of fetal alcohol syndrome. Point is, we have to be careful not to ascribe a genetic origin to every observation.

  • Amanda S

    It seems to me that the bodies of the San are highly adapted to their form of hunting which suggests that this form has continued for many, many generations. Essentially they run down their prey through a combination of tracking skills and using the human ability to keep running long after an antelope or other such species has run out of energy. The San have small, light bodies with well developed buttock muscles. They have the ability to sweat profusely to keep their bodies cool enough. These characteristics make them fantastic long distance runners.

  • Matt

    I was also noticing that the Khoisan have relatively flat faces (low mandibular prognathism), which is more similar to Eurasians and some east african groups, but different than sub-saharan west african groups who tend to have a more prominent prognathism.

    http://95.211.45.61/hanihara.flatness.pdf – c.f. Table 3. Prognathism in the Khoi sample is along the low end for West African samples, along the high end for South East Asian samples (comparable to Malay and Javanese) and around average compared to Amerindian samples. There are other measurements in the above paper where they show more dramatic differences from African and non-African populations.

  • Sandgroper

    “differential retention of ancestral traits of anatomically modern humans” – cool, but why does that happen?

    @25 – We’re not.

  • http://harpending.humanevo.utah.edu Henry Harpending

    #8 and #9:

    I have spent a lot of time with the northwestern folks, apparently those with the least Bantu admixture. Among them there are 3 kinds of animals: !a (meat) is an edible animal, !oma is an inedible animal like a jackal or hyena or European or Black person, and žhu is a person. It is nothing personal, just language, but it is a little bit disconcerting to be referred to as “the !oma” all the time.

    One time a bunch of us pitched camp near a Philippine road construction crew. The guys with me had a lively conversation about the žhu down the road. I asked why they were calling them žhu while I was a !oma. The immediate answer was “Henry, just look at them, they look like real humans.” (The plural of žhu is also žhu but with a high tone.)

    Apparently there is occasionally some reciprocal notice: the movie “The Gods Must be Crazy” was a wild hit in Japan.

    BTW, besides the eyelids, they also have shovel-shaped incisors and appear paedomorphic, supposedly Asian traits. Newborns also have a bluish discoloration at the base of the spine that goes away, the so-called “mongoloid spot”. Make up an adaptive explanation for that!

  • Onur

    Henry,

    I wonder how they classify lighter skinned Mongoloids like Japanese.

  • ackbark

    I have always thought, watching film of Bushmen, that they had an east Asiatic aspect.

    Also, I’ve noticed that a lot of Bushmen have lines across the bridge of their noses,

    http://esaach.org.za/images/thumb/3/37/IMG_4429.JPG/300px-IMG_4429.JPG

    http://www.nairaland.com/347410/picture-thread-1-khoisan-people

    which have always reminded of the lines across the bridge of the nose in bonobos,

    http://www.awf.org/files/4243_image2_bo_3.jpg

    which you don’t see in chimps.

  • ackbark

    What I meant was, does the epicanthic fold evolve out of the lines across the nose?

  • Sandgroper

    ‘Mongolian blue spot’ is interesting, mostly because it seems to defy explanation. My daughter was born with it quite extensively, but fortunately by then I knew what it was and didn’t worry about it, and sure enough it just disappeared with time. Her older male cousin (son of my NE Chinese wife’s sister) also had it extensively. Her brother’s children didn’t have it. Note: I will not buy adaptive explanations like ‘camouflage amongst the rocks’.

    When I have pointed out that some African groups exhibit epicanthic folds, my wife and daughter have both been fascinated to realise that it could be a common ancestral trait that has been differentially retained in East Asians. My explanation to them for the differential retention was ‘maybe founder effect’, but that was a total wild guess.

    @29 – I can see why they would think Filipinos look like real people. I would immediately guess that they would classify East Asians including Japanese as !oma – the overall similarity in stature and appearance is nowhere close.

  • Kosmo

    The underlying structure of the orbit might influence the phenotypic expression of the eyelid. What is an extra fold in the eyelid other than an excess of skin in comparison to the size of the orbit? Perhaps it isn’t just the skin but the size of the underlying bone structure that needs to be looked at.

    I have a fairly extreme example of hooded eyes, and I also happen to have a fairly extreme version of the small, rectangular eye sockets that seemed much more common in paleolithic Europe, before the big aviator sunglasses-looking eye sockets came into vogue there. This is likely just coincidence, but I do wonder if the skin around my eyes might have been pulled into a different shape if my eye sockets were larger.

  • ryan

    #31 those lines across the bridge of the nose that remind you of lines in bonobos … they remind me of lines across the nose of someone spending in a treeless sunny place — they’re just squinting:

    http://www.istockphoto.com/stock-photo-11509296-dramatic-facial-expression-squinting.php

    http://www.123rf.com/photo_10179010_portrait-of-middle-aged-caucasian-man-squinting-in-sunlight.html

  • ackbark

    35. Actually I couldn’t find a really good picture of the lines across the nose. It’s a lot more apparent when you see them speaking on film. The lines appear to go over the nose and into the eyesockets and contribute a lot to general expression.

  • Eurologist

    ‘Mongolian blue spot’ is interesting, mostly because it seems to defy explanation.

    The prenatal environment harbors a strange concoction of hormones. If I were to look into this, I would start with the hypothesis of sexual signalling – with a lot of context lost, over time (exact location on the body, and exact triggering). It seems that once hormones stabilize in the newborn, sooner or later this feature vanishes.

  • http://www.eurasian-sensation.blogspot.com Eurasian Sensation

    One of the theories about the evolution of the epicanthic fold in Asia was that it was an adaptation to snow glare in the Ice Age East Asia. I’m not particularly sold on this idea, but it could also apply to Khoisan, having existed in desert and savannah areas with similar sunlight factors. Same goes for the Nuba linked above.

  • ackbark

    Try squinting at somebody and see how much you love them. I think it’s preserved as a standard of beauty and friendliness.

  • Insightful

    Hey, ackbark. Here’s a picture of an old San Woman and young San girl (possibly her grandaughter). Please note that I don’t see any lines across the nose or nose bridge on the younger one. So your link to ‘bonobos’ is fallacious (at best!) See picture:

    http://i.dailymail.co.uk/i/pix/2012/09/22/article-2207097-15251093000005DC-243_634x702.jpg

  • Onur

    BTW, besides the eyelids, they also have shovel-shaped incisors and appear paedomorphic, supposedly Asian traits. Newborns also have a bluish discoloration at the base of the spine that goes away, the so-called “mongoloid spot”. Make up an adaptive explanation for that!

    I certainly don’t. All those Mongoloid-like traits of Khoisan must surely have been inherited from a common ancestor shared by Mongoloids and Khoisan but not necessarily by the other modern human races. Races are not clades; all sorts of genetic and phenotypic relationships between them are possible.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #41, must surely have been inherited from a common ancestor

    you need to learn how to soften your clauses. they may have been inherited in such a fashion. or, they may be independent occurrences of the same derived trait.

  • Onur

    #42,

    The common Mongoloid-like traits of Mongoloids and Khoisan are very unlikely to be the result of convergent evolution as they are too many in number and very specific in form. Also, there is no apparent reason for their co-evolution independently.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #43, your comment strikes me as phylogenetically uninformed. but perhaps you have an awesome knowledge of phylogenetics which gives you a deep intuition about ancestral trait reconstruction which i’m not privy to. be as that may be, moderate your clauses or i will ban you. i don’t want to mislead my readers, because i am not as convinced by your oracular phylogenetic powers of perception. end of discussion.

  • Onur

    your comment strikes me as phylogenetically uninformed.

    Could you be more explicit?

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #45, you can’t just assess things like this intuitively. you have to construct a tree, and then test the probability of a particular distribution of character states given the tree.

    http://en.wikipedia.org/wiki/Ancestral_reconstruction#Trait_reconstruction

    of course i don’t expect everyone to do something like this, so they need to cautious. more concretely i don’t even believe that the case is persuasive in terms of a cluster of traits on the face of it. there are A LOT of human traits, and you need to be wary of intuitively sampling from the space of characters to validate your own proposition.

  • Onur

    #46,

    In the absence of knowledge about the genetic bases of the Mongoloid-like traits in question, we are bound to try to infer the evolutionary history of the traits in question based on their properties and worldwide distribution (that was what I was doing in my above posts). I certainly don’t want to give the impression that my words are the last word on this matter, sorry if I sounded like that. Only genetics can say the last word. Hopefully, we will learn much about their genetic bases within the next few decades.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    Hopefully, we will learn much about their genetic bases within the next few decades.

    remove the s. i think we’ll know by 2022, so decade.

  • Onur

    remove the s. i think we’ll know by 2012, so decade.

    What is your source? You sound pretty confident.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #49, some of these traits (e.g., epicanthic fold) seem categorical in trait value (i.e., i know east asians where one person in the family does NOT have the trait, all others do). if so, there is a large effect QTL. so all we need are large numbers of cases and controls. within 5-10 years there will be enough public data sets that even people outside of academy can fish around for associations. to give you an explicit example, if you have 1000 korean samples i think you can confidently code them as having epicanthic folds (highest frequency among east asians i think). then you need to a control population, which you have plenty of europeans, west africans, etc. you also have southeast asians, who are an admixed group for this trait. comparisons across these populations on whole-genome data sets by the later years of this decade should be able to ascertain the locus of this trait, and then you can assess how ‘old’ the trait is in these populations. we already have some public bushmen sequences as well.

  • Onur

    #50,

    I have no doubt that we’ll learn much on this matter within a decade. I was confounded because you wrote “I think we’ll know by 2012″ in your previous post. I now assume you meant to say “2022” but made a typo.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    :-) yes.

  • ackbark

    Hey, Insightful,

    here’s The Gods Must Be Crazy,

    https://www.youtube.com/watch?v=f4RYzYgonBg&feature=fvwrel

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at http://www.razib.com

ADVERTISEMENT

See More

ADVERTISEMENT

RSS Razib’s Pinboard

Edifying books

Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »