January 24, 2016

Ancient mtDNA from a Megalithic tomb near Atapuerca

A new study has produced 22 mitochondrial sequences from a Megalithic tomb (dolmen) in Alto del Reinoso, some 7 Km NE of Atapuerca.

Kurt W. Alt, Stephanie Zech et al., A Community in Life and Death: The Late Neolithic Megalithic Tomb at Alto de Reinoso (Burgos, Spain). PLoS ONE 2016. Open access LINK [doi:10.1371/journal.pone.0146176]


The analysis of the human remains from the megalithic tomb at Alto de Reinoso represents the widest integrative study of a Neolithic collective burial in Spain. Combining archaeology, osteology, molecular genetics and stable isotope analysis (87Sr/86Sr, δ15N, δ13C) it provides a wealth of information on the minimum number of individuals, age, sex, body height, pathologies, mitochondrial DNA profiles, kinship relations, mobility, and diet. The grave was in use for approximately one hundred years around 3700 cal BC, thus dating from the Late Neolithic of the Iberian chronology. At the bottom of the collective tomb, six complete and six partial skeletons lay in anatomically correct positions. Above them, further bodies represented a subsequent and different use of the tomb, with almost all of the skeletons exhibiting signs of manipulation such as missing skeletal parts, especially skulls. The megalithic monument comprised at least 47 individuals, including males, females, and subadults, although children aged 0–6 years were underrepresented. The skeletal remains exhibited a moderate number of pathologies, such as degenerative joint diseases, healed fractures, cranial trauma, and a low intensity of caries. The mitochondrial DNA profiles revealed a pattern pointing to a closely related local community with matrilineal kinship patterns. In some cases adjacent individuals in the bottom layer showed familial relationships. According to their strontium isotope ratios, only a few individuals were likely to have spent their early childhood in a different geological environment, whilst the majority of individuals grew up locally. Carbon and nitrogen isotope analysis, which was undertaken to reconstruct the dietary habits, indicated that this was a homogeneous group with egalitarian access to food. Cereals and small ruminants were the principal sources of nutrition. These data fit in well with a lifestyle typical of sedentary farming populations in the Spanish Meseta during this period of the Neolithic.

While the nutritional part has some interest, it is ultimately not too conclusive (high protein diet, similar to that of Dordogne Neolithic, high incidence of caries, three individuals who may have been raised outside the "closely knit" community), so I will focus my attention on the mitochondrial lineages.

These are:
  • U5b - 2
  • U5b2b3 - 4
  • U5b3 - 1*
  • U4 - 1*
  • V - 2
  • K - 4
  • K1a - 1
  • K1a1 - 3
  • X - 3
  • T2b - 3
  • T2a1b - 1*
  • H3 - 1
  • J - 1
The three individuals marked with an asterisk (*) might have been raised in other villages, two of them are adult men and the third one a teen-ager of unknown gender. While the possible immigration of men could suggest matrilocality, the reality is that most individuals buried whose gender could be discerned are men anyhow, and all the rest seem to share the same kind of diet (i.e. probably raised in the Alto del Reinoso community), so most unclear.

Notice also that, because of the limitations in haplogroup identification in ancient DNA, apparent "upstream" lineages such as U5b or K can perfectly be the same as "downstream" ones like U5b2b3 or K1a1 respectively. We just do not know for sure.

My preliminary diagnostic was that it looks a quite typical "mixed Neolithic" pool, much like the one of El Portalón in nearby Atapuerca, with dominance of "Neolithic" lineages (K, X, T, J - maybe also V and H3, not yet detected in Western hunter-gatherers) but also a high frequency of "Paleolithic" ones of the U typology (U5 and U4). It is almost absolutely "pre-modern", lacking the high frequencies (40-60%) of mtDNA H found today (and also found in the Neolithic of Paternabidea and Gurgy, in Navarre and Burgundy respectively) and therefore having great excess of both "Neolithic" and U frequencies.

It must be said that Northern Burgos is even today quite low in mtDNA H, with only 33% of this widespread lineage (half of it H3), being also unusually high (30%) in haplogroup U frequencies (ref.), but in any case the mtDNA pool is at the very least not standard for the wider geography and must have experienced some changes in the meantime therefore.

For that reason I considered comparing with the other nearby sites within this small Atapuerca-La Brújula mountain gates area that divides the Duero from the Ebro basins. The result (in percentages) is as follows:

Site date (BCE) n U5 U4 H* H1 H3 V J T2 X K U3
Reinoso ~3700 22 27 5 - - 5 9 5 18 14 18 -
Portalón ~3000 7 29 - - - 29 - 14 - 14 14 -
Mirador ~2500 20 - - 5 20 - - 10 20 20 20 5

[Ref. links for El Portalón and El Mirador (both in Atapuerca)].

The three sites provide jointly a most interesting sequence for the district. I would say that Reinoso and Portalón seem quite similar, especially considering that the latter sample is very small, allowing for some random fluctuations (decrease in V and T2, increase in H3 and J). However in El Mirador, some 500 years later, we just cannot ignore that there are notable changes.

While the main "Neolithic" lineages (J, T, X, K) remain pretty much the same, all the rest is completely different: U5 (and its faithful sidekick U4) has vanished, has happens with H3 and V, instead we see a sudden outburst of H1 (and H*) and also the less impressive appearance of U3.

What does this mean? Let's go back to the modern mtDNA pool in Northern Burgos Province (n=24) as per Behar 2012:
  • H1: 2 (8%)
  • H3: 4 (17%)
  • U: 8 (33%)
  • K: 2 (8%)
  • T: 2 (8%)
  • J: 2 (8%)
  • Singletons: H*, H4, V, L2
It seems to me that those ancient genetic pools are still very present: although there are greater frequencies of H and U than the average produced by merely admixing Reinoso and Mirador, the underlying typology that we can discern (within H most clearly) seems to conform to what those ancient populations already had. Instead the "Neolithic" lineages are less common.


My guess is that these ancient sites lay all in a key passage of a most strategic route, Spain's National Road 1, St. James' Way, Roman road Ab Asturica Burdigalam and surely much older trade and cultural routes went through that series of mountain passes. This may have attracted more immigrants from the Neolithic settler populations from the Mediterranean, who may have been surrounded by others of more "aboriginal" roots (mostly Paleolithic ascendancy) and rather low visibility for archaeologists.

It remains to be confirmed if Bell Beaker (associated with El Mirador site) may be directly blamed for the introduction of mtDNA H1 in the district. It is indeed possible but in any case the long term impact was limited.

It must also be understood that all this has very little to no relationship with what may have happened in the Basque Country, in spite of being not far away: the genetics involved, both ancient and modern, are quite different. It may be surprising how much the genetic pool can vary in just 200 Km but the rugged geography and diverse ecology seem to favor this kind of sharp distinctions. And, regardless of the causes and the surprise, it may cause, the data is there and is very clear.


  1. This may be evidence of haplogroup K, for example, in Mesolithic Iberian hunter-gatherers (apart from several other ones)? I hardly find it clear:


    1. It is the most confusing text on ancient mtDNA I have ever read, honestly. We have a somewhat more clear Portuguese dataset (Chandler 2005), which tested for specific markers and not a single one of them produced K. K has also not been found in any other European pre-Neolithic sample and becomes instead quite common in the Neolithic. It is also found in Neolithic Syria (contemporary with "Mesolithic" or more properly Epipaleolithic Europe). For all these reasons I think that it is probably a case of misidentification but who knows?

  2. Ok... I remember that you wrote a good critics on that Chandler paper.

  3. Here it is one of your opinions regarding Chandler:


    And I do agree with the following point: «My guess is that these ancient sites lay all in a key passage of a most strategic route, Spain's National Road 1, St. James' Way, Roman road Ab Asturica Burdigalam and surely much older trade and cultural routes went through that series of mountain passes. This may have attracted more immigrants from the Neolithic settler populations from the Mediterranean, who may have been surrounded by others of more "aboriginal" roots (mostly Paleolithic ascendancy) and rather low visibility for archaeologists.»

    It seems that in Iberia, there have been found regions, on which there´s a great deal of evidence that points to a big degree of continuity between the Mesolithic and the Neolithic. There are also some more recent articles about this, that support this idea. I may share it with you... Maybe it would be interesting to join the pieces of the puzzle and write something about it.

    1. Chandler's is still a conference paper with poor exposition and data logs and it has been controversial too. However the main controversy was all the time the presence of mtDNA H in the Epipaleolithic sample, in large numbers BTW. Eventually the discovery of more mtDNA H in pre-Neolithic Iberia and Karelia (several studies) confirmed it. But while the case for mtDNA H has been settled for good (never mind grumpy Manco), the case for mtDNA K has a different origin (your linked extremely confuse paper) and absolutely no confirmation at all. So I'm skeptic in this case, really.

      "There are also some more recent articles about this, that support this idea. I may share it with you..."

      I'm interested of course.

    2. BTW, I just updated the link to Chandler et al. in that old entry, as the old one was dead. Thankfully, Zilhao still keeps it in his portfolio.

  4. Yeah, I´m very skeptical about the mtDNA K finding as well, so I think that the authors should be contacted.

    Nice to see your interest, I hope to get it soon and I´ll post it here.

  5. Maju, sorry to ask, but do you have email? Or better said first, can I send you a personal message?


    1. Of course: lialdamizDELETETHIS@gmail.com (delete the capitalized segment, it's for spam protection).

  6. Maju, I need your help, I just got my Ydna results from Family tree, my haplogroup is G-M201 what does this mean???

    1. G-M201 is just G (http://www.isogg.org/tree/ISOGG_HapgrpG.html). Do you know if you have been tested for downstream markers? The most normal thing for non-Caucasus Europeans is to belong to G2a, which is the best documented lineage arrived in the Neolithic from West Asia (along with E1b-V13 and maybe others).

      Read Wikipedia anyhow: https://en.wikipedia.org/wiki/Haplogroup_G-M201

    2. Also, Iberian Y-DNA frequencies (Adams 2008): http://www.cell.com/AJHG/fulltext/S0002-9297%2808%2900592-2

      G makes up 5% on average but significantly more for Portuguese, Manchegos and Ibicencos.

    3. Hi Maju, thank you so much for your prompt reply, I was almost speechless when I wrote to you, because I got my results like five minutes before I turned to you. It say's my place of origin is Italy, because they found only one identical match in Capua, and everyone else is "one step"away from me. They only have one identical match on my 12 Y-STR values, they found no matches on my 25 and 37 Y-STR values.

      "Do you know if you have been tested for downstream markers?"
      I don't know, how would I find that out?

    4. Yea I already read the wikipedia source, but it just confused me even more. I am very surprised that I don't match the majority of modern Iberians, I am totally exotic looking in every aspect of the meaning Spaniard. I think it's because of my mother's side of the family. This is why I think personally, the woman holds the key to most of our diversity in phenotypes. But this is my own opinion.

    5. It could also be the other way around: that the Capuan lineage has origins where you're from (not sure where exactly but I recall you are Iberian), to make better sense you may want to consider where are the other related lineages, those one-step-removed. Also the fact that you have some STR markers specific to your line probably means that the relation with your Capuan "relative" is rather old (how much, I wouldn't be able to say).

      "I am very surprised that I don't match the majority of modern Iberians".

      Well, you should not. It's not a rare lineage in Iberia at all, and used to be even more common, at least in some areas in ancient times (Neolithic, etc.) Which specific lineage you carry is, from our perspective, random, it could perfectly be something even rarer. Autosomally however, regardless of patrilineage, you should be very similar to your geographic neighbors - but that's another story.

      "I think personally, the woman holds the key to most of our diversity in phenotypes".

      In our individual genomes neither does: as you go back in the genealogy, in time, you realize that your haploid lineages mean very little in the wider picture: it's 100% re. your parents, but 50% of your grandparents legacy, 25% of your great-grandparents, 12.5% of your great-great-grandparents, etc. A couple of centuries ago it just represents a tiny percent, 500 years ago it's absolutely negligible (unless your ancestors were unusually endogamous).

      Another thing is when looking at whole populations, then you may be approximately right but it's just a rule of thumb, a tendency, and there is no strict rule that really applies. For example there was years ago a study on a Colombian port town, where it became apparent that successive waves of Spanish males have made much more impact than the still very dominant Native American matrilineage, whose related autosomal legacy was diluted over and over through the centuries. It may be somewhat exceptional but nonetheless important in order to keep in mind that there's no absolute rule on this.

    6. Just mind boggling to me, I either thought I would of been R1b or E1b1b, but instead I turn out to be G, just in f**** credable.

      "to make better sense you may want to consider where are the other related lineages"

      1 Portugal
      1 Spain
      1 England
      1 Switzerland
      1 Belgium
      1 Netherlands
      1 Bulgaria
      1 Ukraine
      1 Macedonia

    7. That distribution gives the impression of a Balcanic and a West European distribution, without any relevance for Italy, so it's possible that your Italian "relative" is the one more isolated, and hence maybe your particular lineage originated in Iberia or otherwise Western Europe. But extremely tentative anyhow.

    8. Hi L, it seems I'm genetically very rare at least as far as I'm concerned, they haven't even grouped me yet. they only found one identical match on my twelve marker, but none on my 25 and 37 markers. Also as I was looking through my Y-dna haplogroup origins 14 of them are G's, but there's two of them, that are labeled "I M-253" one from Denmark and one UK. This is just getting more and more confusing....

    9. It's just the Y-DNA and anyhow, have you even tested for G2a? Most likely you are G2a, and that is absolutely normal in Europe.

    10. One of the things you have to understand is that what matters is SNP-testing, any STR coincidences you may have with people in other haplogroups are just that: coincidences.

    11. "have you even tested for G2a?"
      No not yet.

      Here's what it say's.

      exact match
      G-M201 Italy 1

      Genetic distance 1 step

      Haplogroup Country Count

      G-CTS4803 Netherlands 1
      G-L140 Macedonia 1
      G-M201 Italy 1
      G-M201 Netherlands 1
      G-M201 Russian Federation 3
      G-M201 Saudi Arabia 1
      G-M201 Spain 2
      G-M201 Ukraine 1
      G-P15 Portugal 1
      G-P15 United Kingdom 1
      G-PF6858 Macedonia 1
      G-U1 Italy 1
      G-Z26414 Switzerland 1
      I-M253 Denmark 1
      I-M253 United Kingdom 1

      Maju, what would you suggest? I want to take my mtDNA test, should I go directly to my mother? or should I take it myself?
      And have you ever taken your Ydna test?
      if not I could give you a full donation for your test, since you posted times are rough for you.

    12. Well, thank for your offer: it's really generous. But I have not the slightest curiosity about my particular Y-DNA, mtDNA or other genetic data. My interest in population genetics is for what it can help us to understand paleohistory in the collective aspect of it, not my own personal genealogy, which anyhow, I know pretty well as far as it matters. Also I am of the strong opinion that, even if we are because our ancestors, we are not them but unique selves. And I like it that way because, to be honest, I don't have good relation with my known ancestors, too conservative (often even fascist) for my taste. The "ancestor" I like the most is not my direct relative but husband of my aunt, almost all the rest just suck.

      Anyhow donations are always welcome. But I will surely spend them in much more important matters for me like dentist, a new pair of glasses, buying a bed, clothes... and getting myself motivated to keep blogging.

    13. As for your doubts...

      1. Almost certainly you are G2a: not only is, by far, the most common G subclade in Europe but also your quasi-matches are often enough within that category (P15 and descendants).

      To test or not to test for G2a and other downstream markers is your decision. I don't know how much it costs nor how much extra info you can get from it. The main use may be to discard less likely lineages like Jewish G1 or other possible (but again unlikely) lines of remote ancestry (read for details: https://en.wikipedia.org/wiki/Haplogroup_G-M201). Personally I think that the most likely meaning of your results is that your very remote patrilineal ancestor arrived to Iberia in the Neolithic (some 7500 years ago) with Cardium Pottery culture but maybe you can get more precise info if you knew your exact subclade. So, if you are truly interested and willing to spend your money and time in finding out all you can, then you should test for all downstream markers possible, as SNPs are much more precise (yes/no answers) than STR matches (of merely approximative value).

      2. You can perfectly test for mtDNA yourself: you should have the same ones as your mother (unless you are unknowingly adopted but that's another kind of investigation). MtDNA evolves very very slowly, so the lineages should be exactly the same unless Fortuna is being extremely capricious and giving you a most unlikely mutation your mother doesn't yet have. This is a "privilege" we men have: we carry both lineages, even if we only transmit the Y-DNA one, if you were a woman trying to find out your Y-DNA roots, you would have to ask your father or brother to get tested instead.

  7. This comment has been removed by the author.

  8. Hi L,I sent you a donation, go get your new glasses and whatever else you can spend the rest on, cheers friend, your time is valuable, so I just wanted to thank you for your time. saludos de um Portugues. You don't have to post this comment keep it private if you'd like, up to you. thanks again.

  9. Hi L, well here it is my friend! I'm confirmed, I belong to haplogroup G L91!! Otzi is my actual genetic relative!! My whole way of thinking just got turned upside down!As I always thought I would belong to R1b, since the majority of Portuguese in my region in northern Portugal are R1b.
    Don't know if I should be excited or dumbfounded.

    1. G2a is not that rare in Iberia: one of every 20 men carries it. And in North Portugal it is the second most common lineage (12%), just after R1b (47%). In fact North Portugal is the most likely place to find a Y-DNA G lineage of all mainland Iberia, just after Ibiza (13%), and followed by Castilla-La Mancha (10%), South Portugal (9%), etc. See: Adams 2008. It would be rare if you'd be Basque, Aragonese, Minorcan or Gascon, but elsewhere it is documented at significant frequencies, averaging 5%.

      Anyhow, it's just one line: by autosomal DNA, what really matters in the end, you are probably most closely related to your geographical neighbors (unless you know of recent immigrant origins that you have never mentioned). In fact it's almost a statement of "North Portugueseness".

      But, yeah, if you want to track the arrival of that lineage to your area, it probably happened in the Neolithic and most likely (my best educated guess) did so via Italy (that's the link to Ötzi: jumping more than 7,000 years back in time!) and South Portugal. You can also track, if you wish, ancestry by that line to Georgia, Turkey and Galilee (maybe even Jesus of Nazareth was your distant "cousin", sure: why not?, if not patrilineal for sure by some other line) and then back to the oldest branch to diverge from the mainline Eurasian patrilineage CF after C - because that is what G is, although the haplogroup probably coalesced much later and we know nothing between the OoA and the founder effect in West Asia probably with the Upper Paleolithic.

      But before your mind explodes, relax: you are for sure more closely related to me than to Ötzi, he is just too old and also he's from another distinct region of Europe. In fact, I'm likely to be a more direct relative of Ötzi (not by patrilineage but by another line) because I have recent ancestry from the Veneto (my grandfather comes from that area and we can track his ancestors, who are also mine, to the 15th century Ferrara).

      So relax: it's just Y-DNA. I know that you're excited but, as "expert" I must throw a jar of cold water: every European, say Gimbutas and Zilhao for example, shares a recent ancestor from the time of Colombus and De Gama. And that also happens worldwide in a slightly deeper time-frame, say Charlemagne's time. So, well, Ötzi is just too old to matter at the personal level.

      Only if you over-emphasize the patrilineage you can say what you say, but I just cannot share that predilection: every line of ancestry weights roughly the same and when we go two centuries back in time it becomes too blurry to matter. That's a reason why I don't personally practice nor recommend personal genetics, except for the research of recent ancestry (adopted people in search of their biological roots, paternity cases and such).

      Another thing is collective genetics: that part is indeed very interesting and informative. But at the personal level, my opinion is that, generally, it is a waste of time and money.

      Citing Winona Ryder's character in "Night on Earth": "As Popeye says: I am who I am". No genealogical research will change that, not for me and not for you either.

  10. Interesting.... I'm T2a1b. I googled my haplogroup and ran into this. This haplotype seems to be everywhere. I've seen it in the Middle East to South America to Northern Europe.

  11. Interesting.... I'm T2a1b. I googled my haplogroup and ran into this. This haplotype seems to be everywhere. I've seen it in the Middle East to South America to Northern Europe.

    1. You can pretty much rest sure that it coalesced in West Asia and spread in the Neolithic to Europe. This is the case quite apparently for the following matrilineages now more or less common in Europe: J, T, X2, W, N1, K, U3, etc. On the other hand we know that at least the following existed in Europe before the Neolithic: H (not necessarily all subclades), U5, U4, U2, U8(xK) and U6 (in spite of being today mostly North African). HV0 and its descendant V also look Paleoeuropean but unconfirmed so far by aDNA and in any case it clearly expanded within Neolithic.


Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).