# DNA analysis and siblings **URL:** \ **Category:** Factual Questions\ **Created:** [January 18, 2018, 2:40pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906 "2018-01-18T14:40:38Z")\ **Posts on this page:** 16\ **Page:** 4
**Author:** ![John\_Mace](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_mace/32/185_2.png) [@John\_Mace](https://boards.straightdope.com/u/John_Mace)\ **Post date:** [January 22, 2018, 11:14pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/61 "2018-01-22T23:14:44Z")
> [@Chronos](#): > > As far as other traits correlating with “race” goes: Two of the four people I linked to share a fairly uncommon mutation affecting their digestive system. Unlike the vast majority of mammals, or even of humans, they’re able to digest lactose as adults. Can you guess, based on what you think their races are, which two they are? Here’s a hint: They happen to be the two tallest ones. The Bantu and the “Caucasian”. The other two come from mostly non-pastoral populations that did not have any evolutionary pressure to favor the mutation for lactose tolerance. We often talk of people being “lactose intolerant”, but that is the basal human condition, and it’s the tolerant folks who are the “mutants”. Yay for mutants!! 🙂 ---
**Author:** ![Riemann](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/riemann/32/3133_2.png) [@Riemann](https://boards.straightdope.com/u/Riemann)\ **Post date:** [January 22, 2018, 11:45pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/62 "2018-01-22T23:45:18Z")
> [@rat\_avatar](#): > > Please show that use with humans in modern science. This is petitio principii (assuming the initial point) as this is exactly why the term was used by 19th and 20th century eugenicists and your Cite calls out this difference as a caveat at the top of the page. I’m perplexed by what you mean here. Obviously any sociological concept of race is, by definition, a social construct. The question at hand is surely whether the biological concept of race is applicable to humans. If you won’t accept that the term “race” is a biological concept _at all_, then obviously there is nothing to discuss. We can’t argue over whether a non-existent concept applies to something. To a geneticist, the concept “race” is defined as in that Wiki article. > [@](#): > > …an informal rank in the taxonomic hierarchy, below the level of subspecies… **genetically distinct populations of individuals within the same species** So the question I’m addressing is whether that concept applies to humans. Is that the question you are talking about, because it’s really not clear from what you have said? ---
**Author:** ![Riemann](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/riemann/32/3133_2.png) [@Riemann](https://boards.straightdope.com/u/Riemann)\ **Post date:** [January 22, 2018, 11:53pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/63 "2018-01-22T23:53:43Z")
> [@John\_Mace](#): > > Sorry, but the logic is lacking. There are biological races, but we can’t say how many there are. Or, we can say how many there are, but the number is arbitrary and two equally good scientist will often come up with completely different classification schemes. If he doesn’t know why race is said to be a social construct, I can only surmise that he doesn’t want to know. It’s a social construct precisely because there is no objective way to divide up human populations in any specific way. That makes any type of division subjective. Hence a social construct, as different societies will define race differently. in fact, the same society might define race differently at different times for political reasons. ??? This just seems to be a paraphrase of what Jerry Coyne explicitly says he agrees with in the first paragraph I quoted. > [@](#): > > The only sense in which they’re right is that the designation of a finite number of easily-distinguished human groups (“races”) is a futile exercise, because we have differentiation within differentiation, making the whole exercise purely subjective. So yes, we’re all in agreement that the historic process of assigning people to one race is arbitrary and invalid. The point that Jerry is making is the “race is social construct” might be taken to mean that there are no genetically distinct subpopulations of humans at all; or that some traits co-segregate. That’s objectively wrong - obviously so, that’s the kind of genetic testing for ancestry that this thread was originally discussing (well, with the caveat that this testing looks mainly at neutral loci, so it doesn’t necessarily follow). ---
**Author:** ![Riemann](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/riemann/32/3133_2.png) [@Riemann](https://boards.straightdope.com/u/Riemann)\ **Post date:** [January 23, 2018, 12:07am UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/64 "2018-01-23T00:07:24Z")
> [@rat\_avatar](#): > > Note the PCA biasing problem I mentioned above, and the small differences here with only 67 populations. I have no idea what your point is. I (and Jerry) are saying nothing more that there are subpopulations of humans with distinct genetic traits; and that some of these traits cosegregate within subpopulations. The populations are obviously not completely distinct from one another with clean breaks. There was always some gene flow. We are all one species. Smaller distinguishable groups that evolved closer to one another will have small differences between them; larger distinguishable groups that evolved further apart may have larger differences. And drawing a circle around any genetically distinguishable part of the human race and calling that part “one race” will always be subjective and arbitrary. Are we actually in disagreement about anything, I’m not sure? ---
**Author:** ![rat\_avatar](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/rat_avatar/32/255_2.png) [@rat\_avatar](https://boards.straightdope.com/u/rat_avatar)\ **Post date:** [January 23, 2018, 2:05am UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/65 "2018-01-23T02:05:09Z")
> [@Riemann](#): > > I have no idea what your point is. I (and Jerry) are saying nothing more that there are subpopulations of humans with distinct genetic traits; and that some of these traits cosegregate within subpopulations. > > The populations are obviously not completely distinct from one another with clean breaks. There was always some gene flow. We are all one species. Smaller distinguishable groups that evolved closer to one another will have small differences between them; larger distinguishable groups that evolved further apart may have larger differences. And drawing a circle around any genetically distinguishable part of the human race and calling that part “one race” will always be subjective and arbitrary. > > Are we actually in disagreement about anything, I’m not sure? You were the ones defending the use of “race” as a valid biological identifier within a generally accepted modern scientific model. The cite that was provided was obviously political in nature, and ignored the null. But to quote the above political opinion cite’s author, > [@](#): > > all scientific progress requires a climate of strong skepticism. > > — J.A. Coyne, The New Republic As he specializes in studying the Speciation fruit fly, Drosophila, it should be trivial to show a paper in which he offers a more rigorous evaluation as to if humans have biological races. As the world’s leading expert on speciation, who made the claim: > [@](#): > > As a biologist, I can’t find any interpretation of that claim that makes much sense. But in the meantime, I think we can recognize the biology behind racial classification while still working to dismantle the bigotry that goes along with it. After all, there are medical, scientific, and evolutionary questions that rest on the genetic structure of our species. Yet he provided a claim that: “You can, for example, distinguish subgroups of “Caucasians” within Europe, distinguishing those of Scandinavian from Italian ancestry simply by their genetic differences.” Yet I can show that only works reliably when you self select the data when referring to the **POPULATIONS** of those areas as a whole. His fault is self selection bias and ignoring the multiple waves of emigration that happened over recent times producing those modern inhabitants. My point is not that there are no differences, but invoking a concept like “race” within these arbitrary boundaries of “Italian” or “Scandinavian” is biased and only valid under this flawed selection model. Consider that “Scandinavian” claim that is even broader than what I offered as an actual example of overlap. Is he including the Sami, the Fenno-Scandinavian, is he ignoring the Fenno-Uric populations that were in the areas like Norrbotten, which was the economic center of Sweden just a few centuries ago? Does it exclude the Finnish, and Kemi Sami from Tornio depite It was the largest merchant town in the North at the time and for some years ranked as the richest town in the Swedish realm in the 16th century just because it doesn’t match “his” definition of what Scandinavian is? Or was the genetic pool of those Finnish and Kemi Sami what he self selected to claim that Scandinavians were “obviously” distinct from Italians. This is the problem with the claim of “race” is that it almost exclusively ignores the realities of the actual genetic diversity of these populations and in place uses arbitrary, non-scientific delineations and actually results in false results that may hide meaningful delineations. To be clear, I have my full SNP sequencing for both mtDNA and yDNA and I actually donated my mtDNA to Genbank because it was different enough to be significant. But this means I also know how unrelated these matches are, even in a fairly recent mutation way to arbitrary political boundaries. And science is lagging because of the assumption that (black, Pacific Islander, white, Hispanic, etc.)… As you are the one claiming that human races are biologically valid groupings, how about enumerating them? How about clarifying how “Hispanic or Latino” relates to genetics at all? Or how about Asian, which includes groups such as the Chukchi & Koryak yet keep Alaska Natives as distinct relating to genetics. From my point of view the argument right now hinges on a single cite, and a definitive, named enumeration of the human “races” is claimed to be scientific while refusing to enumerate those “obvious” segmentation. Why are haplogroups and subclades not sufficient, and far more importantly more accurate? Can you provide any evidence that the eugenics era terms like negroid mongoloid and caucasoid are better tools? More importantly how does eugenics terms like caucasoid actually map to recent data related to haplogroups and subclades? As this is GQ please provide cites, as the only cite that has been offered is one who posits that moving past disproved eugenicist labels to scientifically meaningful descriptions is some left wing conspiracy. “Race” may be applicable to fruit flys, but when it comes to humans, the evidence tends to show that we are just too inbreed for there to be enough diversity to apply that term in a scientific fashion. ---
**Author:** ![rat\_avatar](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/rat_avatar/32/255_2.png) [@rat\_avatar](https://boards.straightdope.com/u/rat_avatar)\ **Post date:** [January 23, 2018, 2:33am UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/66 "2018-01-23T02:33:59Z")
To explain why I said the PSA chart I offered was biased, which is from a eurogenes.blogspot.com.au posting. As I cited before PCA is biased more by sample size than relatedness, - Italian 115 - Han 92 - English 54 - French 49 - Russian 33 - Norwegian 11 - Swedish 2 - Finnish 3 - Sami 0 Yet Finland has some of the highest rates of DNA testing in the world, but it is also fairly genetically isolated, and more importantly would draw all of Scandinavia closer to Russia etc… While the author tries his hardest to be objective, by weighting these samples it pulls Europe tighter and further away from groups which have a common ancestor like the Han, who are over-represented in a way that would accentuate differences. While I have slightly re-purposed the data-set, the author does disagree with a common recent ancestor in between Asia and Finland. The fact that this may be valid is not of matter here, the fact that we all suffer from cognitive biases and need to be as rigorous with our methods as much as possible. While I don’t doubt that assuming the races of the eugenicists reduces the workload, required to write a paper it introduces the effects of cognitive biases that result in a deviation from norm or rationality in judgment as to the real meaning of the data. Without concrete scientific evidence that these arbitrary labels are scientifically significant, it devalues any experiment that utilizes them no matter if it is intentional or just to do a lack of scientific rigor. This is fine if you are trying to sell DNA tests, but is horrible both for science and a society that has very real implications and challenges due to political motivated pseudoscience. That said, I need to dogfood my own posts and apply some Bayesian thinking, and considering the priors, the probability that any argument or cite will be offered to justify the use of or any realistic bounding of the term “race” from a scientific perspective is small. ---
**Author:** ![md2000](https://avatars.discourse-cdn.com/v4/letter/m/73ab20/32.png) [@md2000](https://boards.straightdope.com/u/md2000)\ **Post date:** [January 26, 2018, 4:22pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/67 "2018-01-26T16:22:01Z")
Here’s an interesting detail peripherally related to this thread: > **[Genome Of A Man Born In 1784 Recreated From The DNA Of His Descendants](https://www.techdirt.com/2018/01/25/genome-man-born-1784-recreated-dna-his-descendants/)** > > The privacy implications of collecting DNA are wide-ranging, not least because they don’t relate solely to the person from whom the sample is taken. Our genome is a direct product of our pare… Basically, because he was an outsider in a well-document small pool of DNA, it was easy to determine which one of these is not like the other ones, which DNA does not belong … the interesting detail to me is the unstated suggestion if they got a complete DNA set, that no DNA was lost in dead-end pedigrees. ---
**Author:** ![Tzigone](https://avatars.discourse-cdn.com/v4/letter/t/839c29/32.png) [@Tzigone](https://boards.straightdope.com/u/Tzigone)\ **Post date:** [January 26, 2018, 7:56pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/68 "2018-01-26T19:56:36Z")
> [@septimus](#): > > Why doesn’t Elizabeth Warren just get a DNA test (along with any convenient cousins)? Tests will most likely detect as little as 1% Native American ancestry, right? If not, “Sorry, family history was a myth after all; move on; nothing more to see here.” I have weird one there. My uncle shows as have no Native American DNA on [ancestry.com](http://ancestry.com) DNA test. But I thought he got into the tribe on testing his DNA years ago (is definitely a member). Since he mentioned it being his mother’s mother mother who was Native American (and he knew her as a child) and being able to test because of that, might they have tested mitochondrial DNA, and that shown his heritage, but it already be “diluted” so that 0% showed on [ancestry.com](http://ancestry.com) test? Also, I’ve heard different companies might return different results (they sample populations differently?). I know there was some problem with [Ancesty.com](http://Ancesty.com)’s years ago, but have no idea which company is considered most reliable now. ---
**Author:** ![John\_Mace](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_mace/32/185_2.png) [@John\_Mace](https://boards.straightdope.com/u/John_Mace)\ **Post date:** [January 26, 2018, 11:11pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/69 "2018-01-26T23:11:25Z")
mtDNA doesn’t dilute. IT gets passed, as is, from mother to child. You get mutations over time, but it would be odd to just so happen to find it in one generation. And I don’t think there are any tribes that let you in via DNA testing anyway. Yes, different companies return different results. Some of that is probably on purpose as they want to brag about using proprietary algorithms. ---
**Author:** ![JR\_Brown](https://avatars.discourse-cdn.com/v4/letter/j/d6d6ee/32.png) [@JR\_Brown](https://boards.straightdope.com/u/JR_Brown)\ **Post date:** [January 27, 2018, 12:02am UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/70 "2018-01-27T00:02:37Z")
Regarding the “race” issue, as a practicing biologist: no biologist worth their salt uses “race” for anything other than the sociological meaning. If they want to talk about population groups they call them population groups. Trying to redefine “race” into something that makes biological sense requires completely trashing the sociological definition; for one thing, the entire concept of “black” vs “white” (or vs. anything else for that matter) would have to go immediately. There is no biologically meaningful classification that would put the [!Kung](http://en.wikipedia.org/wiki/%C7%83Kung_people) and the [Yoruba](http://en.wikipedia.org/wiki/Maasai_people) (two African peoples) into the same group and Scandinavians into a separate one.\* Since all of the most distantly-related human groups occur in Africa, **any group that includes all African populations will also include all other humans** , meaning that any biologically-defined concept of “race” would have a whole bunch of different African races and then one big group for “everybody else” - unless you want to have dozens or even hundreds of races, you can’t even split off Scandinavians from Native Americans. Not sure if these are paywalled, but below there are links to a couple of recent papers with attempts to put together human population cladograms. To read these: the branching pattern is the important part. (There are statistical methods for tree-cutting that you would use for a formal analysis, this is the baby-steps version.) 1. If you put two branches in the same group, every branch that can be traced back to the “last common node” that connects them has to go into that same group. (It’s easiest to turn the graphs sideways and think of them like one of those mobiles you hang above a baby’s crib; if you cut the string anywhere, the entire chunk that falls off has to go into one group.) 2. In addition, at minimum, every branch that comes off the “trunk” (i.e., the path from your last common node to the leftmost point of the graph) to the left of your cut point also defines a separate group. So the further to the right the common node between your two starting groups is, the more total groups you will end up with. Figure 1a [here](http://www.nature.com/articles/srep29890) Figure 3a [here](http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1002967) Figure 1b [here](http://science.sciencemag.org/content/319/5866/1100.full) - This is even before considering that there has been a LOT of population mixing since initial divergence, over VERY large geographic and temporal scales, so it’s hard to even define your “fundamental” populations cleanly in the first place. ---
**Author:** ![rat\_avatar](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/rat_avatar/32/255_2.png) [@rat\_avatar](https://boards.straightdope.com/u/rat_avatar)\ **Post date:** [January 27, 2018, 12:24am UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/71 "2018-01-27T00:24:22Z")
> [@Tzigone](#): > > I have weird one there. My uncle shows as have no Native American DNA on [ancestry.com](http://ancestry.com) DNA test. But I thought he got into the tribe on testing his DNA years ago (is definitely a member). Since he mentioned it being his mother’s mother mother who was Native American (and he knew her as a child) and being able to test because of that, might they have tested mitochondrial DNA, and that shown his heritage, but it already be “diluted” so that 0% showed on [ancestry.com](http://ancestry.com) test? > > Also, I’ve heard different companies might return different results (they sample populations differently?). I know there was some problem with [Ancesty.com](http://Ancesty.com)’s years ago, but have no idea which company is considered most reliable now. [ul] [li]Having Native American DNA is different from having Native American ancestry. [/li][li]Most of these “chip” based DNA companies don’t have an extensive NA reference population, or they may not be able to be more specific than the seed population.[/li][li]These chip based STR tests only test a fraction of your DNA, and they look for “interesting” areas in your DNA and only really count the number of “repeats”[/li][li]Your Fathers mother’s mother mother would only could have contributed 12.5% of all DNA and even if he got the whole 12.5% and it wasn’t washed out that small percentage may not have been the DNA with the markers the companies are looking for.[/li][/ul] The autosomal dna you will inherit from your parentage is random. The mtDNA he would have received from the matrilineal lines with minimal changes could have been in haplogroups, A, B, C, D and X. All of which may be mapped to Asia. [![](https://www.familytreedna.com/img/understanding-dna/migration_map_tut.gif) ](https://www.familytreedna.com/img/understanding-dna/migration_map_tut.gif) Documenting haplogroups found within the Native community has been slow for various reasons, including a lack of participants with proven Native heritage on the relevant matrilineal genealogical line. Native American lines also may have disappeared due to institutional efforts and forced sterilization in the 1970s. [ > [@](#): > > …found that one in four American Indian women had been sterilized without her consent. PInkerton-Uri’s research indicated that the Indian Health Service had “singled out full-blooded Indian women for sterilization procedures.” ]([Tribes - Native Voices](https://www.nlm.nih.gov/nativevoices/timeline/543.html)) Forced sterilization by the Indian Health Service impacted \*\*at least \*\*25 percent of Native American women who were between the ages of fifteen and forty-four during the 1970s and these efforts targeted the exact population that is needed to build this data set. That is a huge loss of information, and the women who were subject to or were alive during this forced sterilization tend to not trust the medical profession. ---
**Author:** ![Tzigone](https://avatars.discourse-cdn.com/v4/letter/t/839c29/32.png) [@Tzigone](https://boards.straightdope.com/u/Tzigone)\ **Post date:** [January 27, 2018, 2:42am UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/72 "2018-01-27T02:42:18Z")
> [@John\_Mace](#): > > mtDNA doesn’t dilute. IT gets passed, as is, from mother to child. You get mutations over time, but it would be odd to just so happen to find it in one generation. And I don’t think there are any tribes that let you in via DNA testing anyway. > > Yes, different companies return different results. Some of that is probably on purpose as they want to brag about using proprietary algorithms. Yes, I kniow mtDNA doesn’t dilute. What I mean was was it possible the Native American genes (at least the ones they were looking for) didn’t get passed down through the generations (only half genes passed down each time), so that’s why it didn’t show, even though mitochondrial DNA did. Particularly wondering if that was more likely or it was more likely due to what particular markers they were looking for that caused 0% return. ---
**Author:** ![JKellyMap](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/jkellymap/32/15880_2.png) [@JKellyMap](https://boards.straightdope.com/u/JKellyMap)\ **Post date:** [January 27, 2018, 11:39am UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/73 "2018-01-27T11:39:34Z")
That Jerry Coyne quite from post 51 is just goofy. The first definition he gives for social construct (“unless they mean …”) is much closer to how it is actually used, than the straw man he then proceeds to invent. I say “much closer,’ because even that leaves out some important nuances, elucidated by other Dopers in subsequent posts. ---
**Author:** ![Nava](https://avatars.discourse-cdn.com/v4/letter/n/da6949/32.png) [@Nava](https://boards.straightdope.com/u/Nava)\ **Post date:** [January 27, 2018, 2:15pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/74 "2018-01-27T14:15:11Z")
> [@Chronos](#): > > You’re half-right: Most people would call Trump “Caucasian”, and the other three “Negro”. That’s precisely what “social construct” means: It’s what most people would say. Most people of what culture? In Spain the aboriginals are considered “black skinned but not black”. So, curiously enough, we specify that they are not _negros_. But in the end what happens here is that race happens to be one of those words which mean more than one thing. One of the definitions of race is social (and changes by the culture of the speaker), one is biological, others are cultural, and all of them have changed with time. ---
**Author:** ![Chronos](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/chronos/32/134_2.png) [@Chronos](https://boards.straightdope.com/u/Chronos)\ **Post date:** [January 27, 2018, 2:30pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/75 "2018-01-27T14:30:52Z")
To be clearer, most Americans (at least, those who know that they exist) would consider Australian aboriginals to be of a different race than Africans, too. But we have so little exposure to Australian aboriginals that most Americans, on seeing a picture of one, would incorrectly assume that the pictured person was African, just as **Turble** did. ---
**Author:** ![Batano](https://avatars.discourse-cdn.com/v4/letter/b/b487fb/32.png) [@Batano](https://boards.straightdope.com/u/Batano)\ **Post date:** [January 28, 2018, 12:54pm UTC](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906/76 "2018-01-28T12:54:13Z")
> [@John\_Mace](#): > > There are obvious problems\* differentiating “British” from “Scandinavian”, so it’s not too surprising that you’d see that kind of result. > , > \*Much influx of Scandinavians to the British Isles over the centuries. Starting from the Anglo-Saxon invasion, at least. Remember, it was Anlges, Saxons and Jutes. Jutes are basically Danish. Funny, you don’t look juteish [Previous page](https://boards.straightdope.com/t/dna-analysis-and-siblings/806906.md?page=3)