The Unz Review • An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersiSteve Blog
Genomics of Human Local Adaptation
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeThanksLOLTroll
These buttons register your public Agreement, Disagreement, Thanks, LOL, or Troll with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used three times during any eight hour period.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

From CellPress:

The Genomics of Human Local Adaptation

Jasmin S. Rees, Sergi Castellano and Aida M. Andrés

Modern humans inhabit a variety of environments and are exposed to a plethora of selective pressures, leading to multiple genetic adaptations to local environmental conditions. These include adaptations to climate, UV exposure, disease, diet, altitude, or cultural practice and have generated important genetic and phenotypic differences amongst populations. In recent years, new methods to identify the genomic signatures of natural selection underlying these adaptations, combined with novel types of genetic data (e.g., ancient DNA), have provided unprecedented insights into the origin of adaptive alleles and the modes of adaptation. As a result, numerous instances of local adaptation have been identified in humans. Here, we review the most exciting recent developments and discuss, in our view, the future of this field.

 
Hide 22 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
  1. • Replies: @nokangaroos
    @anon

    Well actually ...
    the list looks sanitized and deliberately uncontroversial -
    fit for Reader´s Digest and the discontented American housewife.

    Just off the top of my head ...

    - The Massai and Samburu lactose tolerance is independent
    - Inhowfar is the Asian ADH (alcohol dehydrogenase) defect (lessened ability to metabolize ethanol) adaptive?
    - No mention whatsoever of Bruce Lahn´s babies (the brain size things)

    Overall I´d say controlled opposition - Cavalli-Sforza signalling virtue.

  2. This is what Greg Cochran and Henry Harpending predicted (The 10,000 Year Explosion, 2010) and what Nicolas Wade worried about (A Troublesome Inheritance, 2014). And it’s a corollary to the data that David Reich discusses (Who We Are And How We Got Here, 2018), notwithstanding its throw-’em-off-the-sled-to-the-pursuing-wolves ending.

    Let’s find out how cancellation-resistant Rees, Castellano and Andrés are.

    • Replies: @Dieter Kief
    @ic1000


    Let’s find out how cancellation-resistant Rees, Castellano and Andrés are.
     
    I first I thought - wow! - Science! - They sure were excited about all their truly amazing new findings.

    Then I felt clumsy because what materialized in my head didn't look so bright & nice any longer.
    Rather creepy stuff.

    I sure wish Rees, Castellano and Andrés (and us...) well!

    , @Anonymous
    @ic1000


    This is what Greg Cochran and Henry Harpending predicted (The 10,000 Year Explosion, 2010) and what Nicolas Wade worried about (A Troublesome Inheritance, 2014).
     
    Please define what “This” is.

    Replies: @ic1000

    , @bomag
    @ic1000


    Let’s find out how cancellation-resistant Rees, Castellano and Andrés are.
     
    TPTB probably don't want too much publicity here; probably just a word to Google to "disappear" the paper.

    But the Narrative loves to celebrate "diversity", so maybe they can spin this as a win for their side!
  3. I wish to point out that some of the mutations included in the list are extremely rare, like the arsenic tolerance of a few natives living in North East Argentina/Atacama region. The great mass of humanity can be divided into three to five classical, homogeneous groups.

    • Disagree: ic1000
    • Replies: @anon
    @J

    I wish to point out that some of the mutations included in the list are extremely rare, like the arsenic tolerance of a few natives living in North East Argentina/Atacama region.

    And yet...they exist.

    The great mass of humanity can be divided into three to five classical, homogeneous groups.

    How do you know that? Why do you believe it?

    Eppur si muove.

  4. @ic1000
    This is what Greg Cochran and Henry Harpending predicted (The 10,000 Year Explosion, 2010) and what Nicolas Wade worried about (A Troublesome Inheritance, 2014). And it's a corollary to the data that David Reich discusses (Who We Are And How We Got Here, 2018), notwithstanding its throw-'em-off-the-sled-to-the-pursuing-wolves ending.

    Let's find out how cancellation-resistant Rees, Castellano and Andrés are.

    Replies: @Dieter Kief, @Anonymous, @bomag

    Let’s find out how cancellation-resistant Rees, Castellano and Andrés are.

    I first I thought – wow! – Science! – They sure were excited about all their truly amazing new findings.

    Then I felt clumsy because what materialized in my head didn’t look so bright & nice any longer.
    Rather creepy stuff.

    I sure wish Rees, Castellano and Andrés (and us…) well!

  5. That list and the abstract are common sensical, not surprising.

    • Agree: Redneck farmer
  6. @ic1000
    This is what Greg Cochran and Henry Harpending predicted (The 10,000 Year Explosion, 2010) and what Nicolas Wade worried about (A Troublesome Inheritance, 2014). And it's a corollary to the data that David Reich discusses (Who We Are And How We Got Here, 2018), notwithstanding its throw-'em-off-the-sled-to-the-pursuing-wolves ending.

    Let's find out how cancellation-resistant Rees, Castellano and Andrés are.

    Replies: @Dieter Kief, @Anonymous, @bomag

    This is what Greg Cochran and Henry Harpending predicted (The 10,000 Year Explosion, 2010) and what Nicolas Wade worried about (A Troublesome Inheritance, 2014).

    Please define what “This” is.

    • Replies: @ic1000
    @Anonymous

    > Please define what “This” is.

    The position that progressives and progressive scientists have staked out is "Race Is A Social Construct."

    To sensible people worldwide, that is obviously true, e.g. a 75% white/25% black person is, socially speaking:
    * "black" in the US, "one drop rule"
    * "Coloured" in apartheid South Africa, mixed-race in current-day Brazil
    * "Quadroon" in Antebellum Louisiana

    What progressives and progressive scientists explicitly mean is "Race Is Only A Social Construct." In other words, there is no meaningful DNA-based difference between races, however defined. There is no way to "scientifically" classify a person as belonging to any given race -- it's merely a question of arbitrary definitions that follow exclusively from social convention. Lewontin's Fallacy is an early example of this position.

    In yet other words, Rousseau uber alles. Nurture and not Nature.

    In Sailer's sensible words, "race" is best thought of as "a partly inbred extended family." Just as "family" can mean different things in different contexts, so to can "race."

    Principal Component Analysis (PCA) of "SNP chip" data is completely consistent with Sailer's perspective. Here is a quick grab of an image that illustrates this point, using data from Li et al's 2008 paper in Science.
    https://pjt111.files.wordpress.com/2011/04/li09pca.jpg

    There are scientific ways to determine a person's racial heritage.

    The paper Steve is discussing by Rees, Castellano and Andrés goes a step further, naming genes with alleles whose frequency varies by race. These alleles confer significant phenotypes onto the people that carry them -- beneficial in certain environments, detrimental in other environments. (That's how natural selection works.)

    These scientific papers are arcane, but their implications are daggers to the heart of the Progressive vision of the way that the world is, and must be.

    Hence, the application of Cancel Culture -- make the intellectuals pay a heavy price for their apostasy.

    Replies: @HA

  7. I wonder about a couple of things. First, where are the psychological adaptations on that list? People in various places have had different lifesstyles. Even as hunter-gatherers, different environments called on different processes, spotting movement, or slight color changes were different different places. People lived in different social structures, some under states, and others in villages. We shoulder, for example, adaptations to low intrasociety violence in the English as opposed to West Africans. Different ones n the Chinese.

    I also wonder about about alleles that started off very rare, and are no now only moderately common. Like, it can take a while for a gene with a 1% selection advantage (pretty big) to go from 1 in 100,000 to 1 in 10. Maybe single base changes happen at the level of a few independent mutation in large populations, but rarer changes will only happen to one person. Like, the modern human population may be large enough to explore every single base mutation, but there is no chance that there are enough of us to explore every possible duplication, inversion, and deletion. They just happen too rarely, and there are too many possibilities.

    I wonder what kind of selection advantage starvation resistance in Samoans had. Did people who had it have 10 times the surviving offspring as those that didn’t? That kind of rapid adaptation can bring bring alleles with some big disadvantages (like sickle cell) especially small populations. In big populations over long periods of time modifier alleles that mitigate the drawbacks of the rapidly evolved adaptation.

    I noticed there were no copy number variants on this list, but they do exist, and are surely adaptations. More research into those would be very interesting. Like we could tell from single base mutations in each copy how many times they happened, how long ago the variants appeared, and estimate what kind of selective advantage they must have had to reach there current frequency. Because genes evolve in parallel, copy number variants give us an idea of what smaller adaptations might do. It would of course not be certain, but when we have no idea what trait a differential key common variant affects, then CNVs give us an idea of what other differences do.

  8. anon[278] • Disclaimer says:
    @J
    I wish to point out that some of the mutations included in the list are extremely rare, like the arsenic tolerance of a few natives living in North East Argentina/Atacama region. The great mass of humanity can be divided into three to five classical, homogeneous groups.

    Replies: @anon

    I wish to point out that some of the mutations included in the list are extremely rare, like the arsenic tolerance of a few natives living in North East Argentina/Atacama region.

    And yet…they exist.

    The great mass of humanity can be divided into three to five classical, homogeneous groups.

    How do you know that? Why do you believe it?

    Eppur si muove.

  9. @ic1000
    This is what Greg Cochran and Henry Harpending predicted (The 10,000 Year Explosion, 2010) and what Nicolas Wade worried about (A Troublesome Inheritance, 2014). And it's a corollary to the data that David Reich discusses (Who We Are And How We Got Here, 2018), notwithstanding its throw-'em-off-the-sled-to-the-pursuing-wolves ending.

    Let's find out how cancellation-resistant Rees, Castellano and Andrés are.

    Replies: @Dieter Kief, @Anonymous, @bomag

    Let’s find out how cancellation-resistant Rees, Castellano and Andrés are.

    TPTB probably don’t want too much publicity here; probably just a word to Google to “disappear” the paper.

    But the Narrative loves to celebrate “diversity”, so maybe they can spin this as a win for their side!

  10. @anon
    https://media1.tenor.com/images/76ebe4c414bae3ed982163a948fa2b6a/tenor.gif

    Replies: @nokangaroos

    Well actually …
    the list looks sanitized and deliberately uncontroversial –
    fit for Reader´s Digest and the discontented American housewife.

    Just off the top of my head …

    – The Massai and Samburu lactose tolerance is independent
    – Inhowfar is the Asian ADH (alcohol dehydrogenase) defect (lessened ability to metabolize ethanol) adaptive?
    – No mention whatsoever of Bruce Lahn´s babies (the brain size things)

    Overall I´d say controlled opposition – Cavalli-Sforza signalling virtue.

  11. Everyone in Europe is descending from Vikings. This information means nothing. It just gets white people all happy about themselves. Good for them.

    There is more diversity in Africa than in the rest of the world put together.

    There is more variety in skin color in Africa than in all the rest of the world put together.

    The San people of South Africa are the most genetically diverse of any living humans studied.

    Just teaching you children about genetic diversity. Listen up! Or it’s the hickory stick fer you!

    • Replies: @anon
    @obwandiyag

    Everyone in Europe is descending from Vikings.

    Lol, no. Not even close.

    There is more diversity in Africa than in the rest of the world put together.

    Lol, no.

    There is more variety in skin color in Africa than in all the rest of the world put together.

    Lol, no! India for a start.

    The San people of South Africa are the most genetically diverse of any living humans studied.

    Lol, you are wrong again.

    Just teaching you children about genetic diversity.

    Perhaps if you stopped wallowing in ignorance you would be less angry. Or at least not so much of an obwindybag.

    Might try it?

  12. Today is Marcus Garvey’s birthday. Let’s have a toast! Few people understood “local adaptation” as well as he. Or at least pretended to.

    It is also the birthday of Johnny Mercer, who, like Mick Jagger and Keith Richards, was born in Dartford. However, before you take “one more for the road”, we’re talking about this Johnny Mercer:

    https://www.gov.uk/government/people/johnny-mercer

    (What the bloody hell does “Minister for Defence People” mean?)

    Margaret Thatcher attempted unsuccessfully to represent Dartford at the beginning of her parliamentary career. Mrs Thatcher does have something in common with Mr Garvey– both died in London.

  13. @Anonymous
    @ic1000


    This is what Greg Cochran and Henry Harpending predicted (The 10,000 Year Explosion, 2010) and what Nicolas Wade worried about (A Troublesome Inheritance, 2014).
     
    Please define what “This” is.

    Replies: @ic1000

    > Please define what “This” is.

    The position that progressives and progressive scientists have staked out is “Race Is A Social Construct.”

    To sensible people worldwide, that is obviously true, e.g. a 75% white/25% black person is, socially speaking:
    * “black” in the US, “one drop rule”
    * “Coloured” in apartheid South Africa, mixed-race in current-day Brazil
    * “Quadroon” in Antebellum Louisiana

    What progressives and progressive scientists explicitly mean is “Race Is Only A Social Construct.” In other words, there is no meaningful DNA-based difference between races, however defined. There is no way to “scientifically” classify a person as belonging to any given race — it’s merely a question of arbitrary definitions that follow exclusively from social convention. Lewontin’s Fallacy is an early example of this position.

    In yet other words, Rousseau uber alles. Nurture and not Nature.

    In Sailer’s sensible words, “race” is best thought of as “a partly inbred extended family.” Just as “family” can mean different things in different contexts, so to can “race.”

    Principal Component Analysis (PCA) of “SNP chip” data is completely consistent with Sailer’s perspective. Here is a quick grab of an image that illustrates this point, using data from Li et al’s 2008 paper in Science.

    There are scientific ways to determine a person’s racial heritage.

    The paper Steve is discussing by Rees, Castellano and Andrés goes a step further, naming genes with alleles whose frequency varies by race. These alleles confer significant phenotypes onto the people that carry them — beneficial in certain environments, detrimental in other environments. (That’s how natural selection works.)

    These scientific papers are arcane, but their implications are daggers to the heart of the Progressive vision of the way that the world is, and must be.

    Hence, the application of Cancel Culture — make the intellectuals pay a heavy price for their apostasy.

    • Replies: @HA
    @ic1000

    "There are scientific ways to determine a person’s racial heritage."

    Yes and no. The problem is in the various clustering algorithms. More to the point, even if you could agree on which one to use, there's no scientific way to determine whether 5 races or 11 (or 3, or even 1, for that matter) is the scientifically optimal choice. Obviously, the choice of 11 will reduce the overall error, but that's a general feature of increasing the degrees of freedom in any kind of data analysis (and which just leads to overfitting if taken too far -- note that the same people that claim there's only 1 race would be just as happy saying there are 7.5 billion of them). Moreover, the proposed adjustments for increasing the degrees of freedom (e.g. the Bayes or Akaike information criterion, adjusted R-squared, etc.) have limited usefulness in the real world, at least in my experience.

    Similarly, we've come to rely on 95% confidence intervals as the knee-jerk gold standard -- too much so, many would argue, since they're just as arbitrary.

    Replies: @ic1000

  14. @ic1000
    @Anonymous

    > Please define what “This” is.

    The position that progressives and progressive scientists have staked out is "Race Is A Social Construct."

    To sensible people worldwide, that is obviously true, e.g. a 75% white/25% black person is, socially speaking:
    * "black" in the US, "one drop rule"
    * "Coloured" in apartheid South Africa, mixed-race in current-day Brazil
    * "Quadroon" in Antebellum Louisiana

    What progressives and progressive scientists explicitly mean is "Race Is Only A Social Construct." In other words, there is no meaningful DNA-based difference between races, however defined. There is no way to "scientifically" classify a person as belonging to any given race -- it's merely a question of arbitrary definitions that follow exclusively from social convention. Lewontin's Fallacy is an early example of this position.

    In yet other words, Rousseau uber alles. Nurture and not Nature.

    In Sailer's sensible words, "race" is best thought of as "a partly inbred extended family." Just as "family" can mean different things in different contexts, so to can "race."

    Principal Component Analysis (PCA) of "SNP chip" data is completely consistent with Sailer's perspective. Here is a quick grab of an image that illustrates this point, using data from Li et al's 2008 paper in Science.
    https://pjt111.files.wordpress.com/2011/04/li09pca.jpg

    There are scientific ways to determine a person's racial heritage.

    The paper Steve is discussing by Rees, Castellano and Andrés goes a step further, naming genes with alleles whose frequency varies by race. These alleles confer significant phenotypes onto the people that carry them -- beneficial in certain environments, detrimental in other environments. (That's how natural selection works.)

    These scientific papers are arcane, but their implications are daggers to the heart of the Progressive vision of the way that the world is, and must be.

    Hence, the application of Cancel Culture -- make the intellectuals pay a heavy price for their apostasy.

    Replies: @HA

    “There are scientific ways to determine a person’s racial heritage.”

    Yes and no. The problem is in the various clustering algorithms. More to the point, even if you could agree on which one to use, there’s no scientific way to determine whether 5 races or 11 (or 3, or even 1, for that matter) is the scientifically optimal choice. Obviously, the choice of 11 will reduce the overall error, but that’s a general feature of increasing the degrees of freedom in any kind of data analysis (and which just leads to overfitting if taken too far — note that the same people that claim there’s only 1 race would be just as happy saying there are 7.5 billion of them). Moreover, the proposed adjustments for increasing the degrees of freedom (e.g. the Bayes or Akaike information criterion, adjusted R-squared, etc.) have limited usefulness in the real world, at least in my experience.

    Similarly, we’ve come to rely on 95% confidence intervals as the knee-jerk gold standard — too much so, many would argue, since they’re just as arbitrary.

    • Replies: @ic1000
    @HA

    HA, thanks for the added color -- you know much more than I do. But I think the most basic point is that many or most PCA-based analyses contradict the high-decibel claims that "race is only a social construct, no biology involved."

    Worth emphasizing that it is prestige scientists and science journalists who are behind the megaphone on this, such as Adam Rutherford and Angela Saini (recent Sailer post on the topic).

    [Edit: what follows is my effort to point out to interested readers where HA and I likely agree.]

    In the PCA figure I showed in comment #13 supra, each dot represents a person of known racial pedigree. Only people who have 4/4 grandparents (or 8/8 great-grandparents) of the same pedigree are shown. By excluding inter-group marriages (which became much more practical and common at the start of the Steamship/Railroad Age), clarity on the structure of the ancestral in-breeding groups ("races"/"ethnies"/"population groups", as you please) is maximized.

    That PCA diagram is one of thousands that has appeared in the biological literature in the past two decades. The relationships between and among groups is highly reproducible. The outcomes are robust: they don't change much based on "which particular individuals were genotyped" or "which SNP chip or sequencing procedure was selected" or "which particular mathematical protocol was used."

    In the comment #13 figure, African individuals (red dots) cluster together. As do Europeans (green). And East Asians (light brown).

    As your comment notes, these graphs -- and reality -- shows some clear distinct-group features and some clinal features. That underlies most of the scientifically meaningful controversy. E.g. while Oceanea (blue dots) is clearly separate from Middle East (dark brown), East Asia (light brown) and America (purple) intersect. Instead of being classified as two groups, should they be a single East Asia/America group?

    That is an "interesting" pedagogical and political question. In a better world, the answer would hinge on what is to be accomplished. Improving medical diagnostics or pleasing Ancestry.com customers or profiling rapists from semen or understanding ancient migration patterns -- these different goals can lead to different lumper/splitter decisions and different naming conventions.

    But the underlying reality of the PCA figure remains the same.

    Replies: @HA

  15. If you’ve ever wondered why you often see blacks wear coats and knit caps on remarkably nice days, this paper on cold perception gene TRPM8 is interesting. https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1007298. Basically, the prevalence of the gene has a very strong connection with a population’s latitude. It’s somehow involved in regulating the body’s perception and response to cold, though exactly what it does is unclear.

    The upstream variant rs10166942 shows extreme population differentiation, with frequencies that range from 5% in Nigeria to 88% in Finland

    Interestingly, it’s associated with migraines as well.

  16. Interesting and highly topical throw-away line in the paper cited by Steve:

    Introgressed alleles [i.e. acquired from Neanderthals and Denisovans] have been suggested to be particularly important in resistance against viruses, especially against RNA viruses in Europeans [44].

    • Replies: @anon
    @Eagle Eye

    More in depth here from 2019.

    Evidence that RNA viruses drove of adaptive introgression between Neanderthals and modern humans

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6176737/

  17. @HA
    @ic1000

    "There are scientific ways to determine a person’s racial heritage."

    Yes and no. The problem is in the various clustering algorithms. More to the point, even if you could agree on which one to use, there's no scientific way to determine whether 5 races or 11 (or 3, or even 1, for that matter) is the scientifically optimal choice. Obviously, the choice of 11 will reduce the overall error, but that's a general feature of increasing the degrees of freedom in any kind of data analysis (and which just leads to overfitting if taken too far -- note that the same people that claim there's only 1 race would be just as happy saying there are 7.5 billion of them). Moreover, the proposed adjustments for increasing the degrees of freedom (e.g. the Bayes or Akaike information criterion, adjusted R-squared, etc.) have limited usefulness in the real world, at least in my experience.

    Similarly, we've come to rely on 95% confidence intervals as the knee-jerk gold standard -- too much so, many would argue, since they're just as arbitrary.

    Replies: @ic1000

    HA, thanks for the added color — you know much more than I do. But I think the most basic point is that many or most PCA-based analyses contradict the high-decibel claims that “race is only a social construct, no biology involved.”

    Worth emphasizing that it is prestige scientists and science journalists who are behind the megaphone on this, such as Adam Rutherford and Angela Saini (recent Sailer post on the topic).

    [Edit: what follows is my effort to point out to interested readers where HA and I likely agree.]

    In the PCA figure I showed in comment #13 supra, each dot represents a person of known racial pedigree. Only people who have 4/4 grandparents (or 8/8 great-grandparents) of the same pedigree are shown. By excluding inter-group marriages (which became much more practical and common at the start of the Steamship/Railroad Age), clarity on the structure of the ancestral in-breeding groups (“races”/”ethnies”/”population groups”, as you please) is maximized.

    That PCA diagram is one of thousands that has appeared in the biological literature in the past two decades. The relationships between and among groups is highly reproducible. The outcomes are robust: they don’t change much based on “which particular individuals were genotyped” or “which SNP chip or sequencing procedure was selected” or “which particular mathematical protocol was used.”

    In the comment #13 figure, African individuals (red dots) cluster together. As do Europeans (green). And East Asians (light brown).

    As your comment notes, these graphs — and reality — shows some clear distinct-group features and some clinal features. That underlies most of the scientifically meaningful controversy. E.g. while Oceanea (blue dots) is clearly separate from Middle East (dark brown), East Asia (light brown) and America (purple) intersect. Instead of being classified as two groups, should they be a single East Asia/America group?

    That is an “interesting” pedagogical and political question. In a better world, the answer would hinge on what is to be accomplished. Improving medical diagnostics or pleasing Ancestry.com customers or profiling rapists from semen or understanding ancient migration patterns — these different goals can lead to different lumper/splitter decisions and different naming conventions.

    But the underlying reality of the PCA figure remains the same.

    • Replies: @HA
    @ic1000

    I agree with all that. I'll highlight the following:

    "In a better world, the answer would hinge on what is to be accomplished. Improving medical diagnostics or pleasing Ancestry.com customers or profiling rapists from semen or understanding ancient migration patterns "

    More generally, you might look at which classification strategy is most successful, as opposed to arguing which is theoretically more correct. For example, in finance -- let's take the particular case of a trading strategy -- there are usually more practical ways of determining which one is best (say, Sharpe ratio) than some high-faluting Akaike Information criterion (which is not to say that those more practical ways are necessarily any less arbitrary). Anyway, you'd look at which classification scheme gives you the strategy with the most profits over some year or quarter and then assume (again, completely arbitrarily) that whatever wins is the best overall (even though the winner may have just been lucky during that particular time period). As noted earlier, using more categories or subdivisions (or "races") can sometimes lead to higher profits because you're able to describe things with more detail, but it can also result in a model that is overfitted and therefore does more poorly because you have less data for each classification.

    Same thing goes with societies, be they human, or gopher, or bee, or whatever. The lumpers and the splitters (and more generally, the ones who want to subdivide others into few groups vs. those who prefer larger numbers) will each see advantages and disadvantages to their approach in forming kinship bonds, finding "appropriate" mates, or identifying enemies. In a given century, one or the other strategy might have been a clear winner or loser, but that doesn't have to always be true, and in the long-term, it's anyone's guess.

    Again, science can only get you so far, and we need to recognize and admit what is arbitrary and ad-hoc (e.g., whether a 95% confidence interval is truly better or worse than a 99% confidence interval).

  18. @ic1000
    @HA

    HA, thanks for the added color -- you know much more than I do. But I think the most basic point is that many or most PCA-based analyses contradict the high-decibel claims that "race is only a social construct, no biology involved."

    Worth emphasizing that it is prestige scientists and science journalists who are behind the megaphone on this, such as Adam Rutherford and Angela Saini (recent Sailer post on the topic).

    [Edit: what follows is my effort to point out to interested readers where HA and I likely agree.]

    In the PCA figure I showed in comment #13 supra, each dot represents a person of known racial pedigree. Only people who have 4/4 grandparents (or 8/8 great-grandparents) of the same pedigree are shown. By excluding inter-group marriages (which became much more practical and common at the start of the Steamship/Railroad Age), clarity on the structure of the ancestral in-breeding groups ("races"/"ethnies"/"population groups", as you please) is maximized.

    That PCA diagram is one of thousands that has appeared in the biological literature in the past two decades. The relationships between and among groups is highly reproducible. The outcomes are robust: they don't change much based on "which particular individuals were genotyped" or "which SNP chip or sequencing procedure was selected" or "which particular mathematical protocol was used."

    In the comment #13 figure, African individuals (red dots) cluster together. As do Europeans (green). And East Asians (light brown).

    As your comment notes, these graphs -- and reality -- shows some clear distinct-group features and some clinal features. That underlies most of the scientifically meaningful controversy. E.g. while Oceanea (blue dots) is clearly separate from Middle East (dark brown), East Asia (light brown) and America (purple) intersect. Instead of being classified as two groups, should they be a single East Asia/America group?

    That is an "interesting" pedagogical and political question. In a better world, the answer would hinge on what is to be accomplished. Improving medical diagnostics or pleasing Ancestry.com customers or profiling rapists from semen or understanding ancient migration patterns -- these different goals can lead to different lumper/splitter decisions and different naming conventions.

    But the underlying reality of the PCA figure remains the same.

    Replies: @HA

    I agree with all that. I’ll highlight the following:

    “In a better world, the answer would hinge on what is to be accomplished. Improving medical diagnostics or pleasing Ancestry.com customers or profiling rapists from semen or understanding ancient migration patterns “

    More generally, you might look at which classification strategy is most successful, as opposed to arguing which is theoretically more correct. For example, in finance — let’s take the particular case of a trading strategy — there are usually more practical ways of determining which one is best (say, Sharpe ratio) than some high-faluting Akaike Information criterion (which is not to say that those more practical ways are necessarily any less arbitrary). Anyway, you’d look at which classification scheme gives you the strategy with the most profits over some year or quarter and then assume (again, completely arbitrarily) that whatever wins is the best overall (even though the winner may have just been lucky during that particular time period). As noted earlier, using more categories or subdivisions (or “races”) can sometimes lead to higher profits because you’re able to describe things with more detail, but it can also result in a model that is overfitted and therefore does more poorly because you have less data for each classification.

    Same thing goes with societies, be they human, or gopher, or bee, or whatever. The lumpers and the splitters (and more generally, the ones who want to subdivide others into few groups vs. those who prefer larger numbers) will each see advantages and disadvantages to their approach in forming kinship bonds, finding “appropriate” mates, or identifying enemies. In a given century, one or the other strategy might have been a clear winner or loser, but that doesn’t have to always be true, and in the long-term, it’s anyone’s guess.

    Again, science can only get you so far, and we need to recognize and admit what is arbitrary and ad-hoc (e.g., whether a 95% confidence interval is truly better or worse than a 99% confidence interval).

    • Thanks: ic1000
  19. @Eagle Eye
    Interesting and highly topical throw-away line in the paper cited by Steve:

    Introgressed alleles [i.e. acquired from Neanderthals and Denisovans] have been suggested to be particularly important in resistance against viruses, especially against RNA viruses in Europeans [44].
     

    Replies: @anon

    More in depth here from 2019.

    Evidence that RNA viruses drove of adaptive introgression between Neanderthals and modern humans

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6176737/

  20. anon[334] • Disclaimer says:
    @obwandiyag
    Everyone in Europe is descending from Vikings. This information means nothing. It just gets white people all happy about themselves. Good for them.

    There is more diversity in Africa than in the rest of the world put together.

    There is more variety in skin color in Africa than in all the rest of the world put together.

    The San people of South Africa are the most genetically diverse of any living humans studied.

    Just teaching you children about genetic diversity. Listen up! Or it's the hickory stick fer you!

    Replies: @anon

    Everyone in Europe is descending from Vikings.

    Lol, no. Not even close.

    There is more diversity in Africa than in the rest of the world put together.

    Lol, no.

    There is more variety in skin color in Africa than in all the rest of the world put together.

    Lol, no! India for a start.

    The San people of South Africa are the most genetically diverse of any living humans studied.

    Lol, you are wrong again.

    Just teaching you children about genetic diversity.

    Perhaps if you stopped wallowing in ignorance you would be less angry. Or at least not so much of an obwindybag.

    Might try it?

  21. I actually know something about this and you are just a child with your initials for things. Prove everything I said is wrong. Ha, you can’t. You don’t even know where I got it. Cretin.

    • Agree: ic1000
  22. I am demonstrably right. You are so stupid you are not even wrong.

Comments are closed.

Subscribe to All Steve Sailer Comments via RSS