Failure to replicate sex bias in steppe migrations

Angela

Elite member
Messages
21,823
Reaction score
12,325
Points
113
Ethnic group
Italian
See: Reich, Lazaridis
http://biorxiv.org/content/biorxiv/early/2017/03/14/114124.full.pdf

"Failure to Replicate a Genetic Signal for Sex Bias in the Steppe Migration into Central Europe"

The reference is to this paper:
Amy Goldberg et al
Familial migration of the Neolithic contrasts massive male migration during Bronze Age inEurope inferred from ancient X chromosomes
http://biorxiv.org/content/biorxiv/early/2016/09/30/078360.full.pdf

and discussed by us here:
http://www.eupedia.com/forum/thread...inated?highlight=chromosome+steppe+migrations

This is how science is supposed to work. I guess those of us who doubted this whole huge harems idea had a point.

From the Reich, Lazaridis document:

"Goldberg et al.(1) used genome-wide ancient DNA data (2) from central European Bronze Age (BA)populations, and their three ancestral sources of steppe pastoralists (SP), Anatolian farmers (AF), andEuropean hunter-gatherers (HG), to investigate whether the SP migration into central Europe after5,000 years ago (3, 4) was sex biased. By estimating a lower proportion of SP ancestry on the Xchromosome(36.6%) which is primarily carried in females than on the autosomes (61.8%), theysuggested that the migration involved a ratio of 5-14 SP males for every female...

We attempted to replicate this finding using qpAdm (3), which leverages allele frequency correlationsbetween the admixed (BA) and source (SP, AF, HG) populations with distant outgroups to eliminatepotential biases due to genetic drift between the true source populations and the ones used assurrogates for them...

For the BA population, we estimate 61.4±2.9% SP, 31.0±1.2% AF,7.6±2.9% HG ancestry using all autosomal SNPs and 67.5±17% SP, 26.5±6.9% AF, 6.0±16.4% HGusing all X-chromosome SNPs; thus we do not find less SP ancestry on the X-chromosome.


I find it interesting they're using "SP", steppe people, perhaps, instead of Yamnaya or Indo-Europeans etc.
 
This dual opposition is a bit simplistic. It depends what period and region we are looking at. It has always been clear that the Corded Ware expansion was conducted both by men and women. The very first, pre-Yamna incursions of R1b into the Balkans were led predominantly by men. There may have been other groups of male adventurers spreading all over central and western Europe during the Late Neolithic (Bell Beaker culture). But during the Late Yamna there seem to have been a real family migration from the Steppe to the Hungarian plain, which then continued up to Czechia, Poland and Germany (Unetice culture). The fateful period for the expansion of R1b lineages was 2500 to 2000 BCE. This is when we see in an explosion of lineages under P312 and U106 in the R1b phylogeny. In just a few generations over a hundred new subclades appeared, from which almost half of modern West Europeans descend. This could only have happened if R1b men took numerous wives. And indeed West Europeans have overwhelmingly Neolithic mtDNA, in contrast to central and eastern Europeans, who have a considerable percentage of Steppe mtDNA. I don't have the exact data because only deep mtDNA subclades can differentiate Steppe from Neolithic H, J, K and T, and WHG/SHG U5a2 and U5b from EHG U5a1. But if we take only haplogroups I, U4 and W as a fraction of Steppe maternal ancestry, here is what we get:

- Spain : 4.4% of I, U4 and W
- Denmark : 5.4%
- Portugal : 5.7%
- Switzerland : 5.8%
- France : 6.4%
- Norway : 6.4%
- Ireland : 6.6%
- Germany : 6.7%
- Albania : 6.8%
- England : 7.4%
- Scotland : 7.5%
- Bulgaria : 7.8%
- Croatia : 8.3%
- Belarus : 8.5%
- Lithuania : 9%
- Ukraine : 10.1%
- Hungary : 10.4%
- Slovakia : 10.4%
- Poland : 10.7%
- Karachay-Balkars : 11.3%
- Serbia : 13.6%
- Dargins : 13.9%
- Bashkirs : 15.9%
- Avars : 16.2%
- Chuvashs : 18.2%

It's not really correct because some U4 was WHG and W1 was Neolithic, so the numbers are a bit higher than they should be for Steppe mtDNA in Western Europe.

What's interesting is that the British Isles do have a few percents more Steppe mtDNA than the rest of Western Europe.


I am not sure of the location and time frame of the samples analysed by the two teams, but we should expect considerable differences between the Late Neolithic period in Central Europe (Bell Beaker period), and the Early Bronze Age (Unetice period). Furthermore, Central Europe is a vague term, and northern Corded Ware sites would have considerably more female Steppe ancestry than any contemporary Steppe culture in the southern half of the continent (south of Czechia and Slovakia).
 
From the Goldberg et al paper:
"We analyzed published (6) ancient samples that have been genotyped for a set of 1,240,000 SNPs, including 49,711 on the X chromosome. Under notation from (6), for the early Neolithic migration from Anatolia, we considered individuals from the CEM population label for ‘selection label 2’; for the Late Neolithic/Bronze Age migrations from the Pontic Steppe, we considered individuals with ‘archeological culture’ label Central_LNBA. These subsets of the data geographically restrict analyses to Central Europeans, decreasing potential variation from spatial variation within Europe. Additionally, while the samples each span approximately one thousand years, the small correlations (< 0.1) between X or autosomal ancestry and calibrated dates are not statistically significant."

This specific information about the samples being analyzed comes from the Reich, Lazaridis rebuttal:


Reich Lazaridis rebuttal of Amy Goldberg et al.PNG

I just quickly skanned the entry at Jean Manco's ancient dna page. The steppe populations seem to have carried more than U4, W, and X to the best of my recollection. Is that incorrect?

If someone has the information readily to hand about the cultural affiliation of these samples it would be interesting to see it, as well as the mtDna.
 
We attempted to replicate this finding using qpAdm (3), which leverages allele frequency correlationsbetween the admixed (BA) and source (SP, AF, HG) populations with distant outgroups

That's a red flag. As far as I know using distant outgroups is never a good idea. If you use only Asian and African outgroups then there's hardly any difference between EEF and CHG for example.

EDIT: Never mind. Here are their outgroups; Mota (5), Ust_Ishim (6), Kostenki14 (7), GoyetQ116-1 (7),Vestonice16 (7), MA1(8), AfontovaGora3 (7), and Levantine Neolithic farmers (9).

Those are good outgroups for determining Steppe, EEF, WHG ancestry.
 
I just quickly skanned the entry at Jean Manco's ancient dna page. The steppe populations seem to have carried more than U4, W, and X to the best of my recollection. Is that incorrect?

If someone has the information readily to hand about the cultural affiliation of these samples it would be interesting to see it, as well as the mtDna.

Steppe
BA Central Europe
 
@Maciamo,

It's difficult to determine Steppe mtDNA frequencies because....

>Natural selection
>Steppe shared some mtDNA with EEF and Eastern WHG.
>Modern and ancient samples haven't been tested at a high enough coverage to find many Steppe mHGs.

T1a and hg I were popular in both Yamnaya and Andornovo. T1a and I haven't been found in Neolithic Spain. Modern Spain and Lithuania have about the same amount of T1a and I. That doesn't make sense right? There's also hardly any variation of U2e frequencies in Europe. Forces besides maternal steppe ancestry affect T1a, I, W6, U2e, etc. frequencies.

U5a correlates best with Steppe ancestry because it's the only mHG that was popular in Steppe, unpopular in EEF, and has been tested in many modern samples. But Eastern HGs also had a lot of U5a and some Hungarian farmers had a decent amount. So some U5a in Eastern Europe is certainly of non Steppe origin.
 
They argue the older study concluded Steppe/MN admixture was sex bias becuase the ADMIXTURE method they used overestimates Steppe ancestry in the autosomes.

Supervised ADMIXTURE predicts real ancestry poorly in this setting (Figure 1b). The estimation error (estimated – real ancestry) is strongly correlated (r=0.91) with theestimated SP ancestry, allowing us to predict it by a regression (Figure 1c), which indicates upwardbias for high SP ancestry estimates and downward bias for low ones. For the pool of 16 individualsthat the authors of (1) estimated had 36.6% ancestry, the error is predicted to be -19.5%, largelyeliminating the discrepancy between the X-chromosome estimate of (1) and the autosomal ancestryestimates of ADMIXTURE (1) and qpAdm. These results show that bias in the estimation ofadmixture proportions, rather than sex bias in the steppe migration, drives the findings of (1).
 
Good stuff. I argued before that with such large scale population replacement in Central North and Central West Europe, it would be impossible to explain with only male invasion.
 
Forces besides maternal steppe ancestry affect T1a, I, W6, U2e, etc. frequencies.

W6 is also not exclusively Steppe (but W6a probably is).

There are only four ancient W6 samples I'm aware of:

W6: Va7, ~5500-4500 BC, Vinča culture Romania

W6c: Lopatino I, ~3090-2910 BC, Yamnaya Russia

W6a: Lopatino II, ~3500-2700 BC, Yamnaya Russia
W6a: Esperstedt, ~2566-2477 BC, Corded Ware Germany
 
So there were not ten farmer females trying to chase together a steppe warrior with blond hair weaving freely on the wind as he was driving his chariot with a naked and musculed chest as I think some Yamnayists were dreaming about?
 
So there were not ten farmer females trying to chase together a steppe warrior with blond hair weaving freely on the wind as he was driving his chariot with a naked and musculed chest as I think some Yamnayists were dreaming about?

Lol!! Ten to one sounds crazy, I don't know how they came up with that. I do think though that Steppe/MN admixture was sex bias to some extent because of Y DNA R1 frequencies and frequencies of typical EEF mtDNA. But when looking at BA Central European mtDNA it's obvious many Steppe women migrated alongside their men.
giphy.gif
 
From my blog post...

Corded Ware. N=67.
mtDNA: Steppe 42% (28), Farmer 41% (27), Unknown 17% (12)
Y DNA: 100% Steppe.
Autosomal: ~75% Steppe, ~25% Farmer


Bell Beaker. N=42.
Steppe=38% (16), Farmer=36% (16), Unknown 26% (10)
Y DNA: 100% Steppe
Autosomal: ~50% Steppe, ~50% Farmer


Unetice. N=94
Steppe=36% (34), Farmer=28% (26), Unknown 36% (34)
Y DNA: 100% Farmer?(yHG I2, could be Steppe aswell)
Autosomal: ~60% Steppe, ~40% Farmer.
 
So there were not ten farmer females trying to chase together a steppe warrior with blond hair weaving freely on the wind as he was driving his chariot with a naked and musculed chest as I think some Yamnayists were dreaming about?
Even funnier, the fact that they weren't blonde but looked more like northern West Asians via skin color.
 
Even funnier, the fact that they weren't blonde but looked more like northern West Asians via skin color.

But besides color they definitely looked distinctively European. There's more than a color difference between West Asians and Europeans.
 
Even funnier, the fact that they weren't blonde but looked more like northern West Asians via skin color.

I think it is not a problem at all, you know, in the steppe there were many corners and there were Asian/Armenian R1b, CW R1a, and R1b-L51, with blond, black and even green hair if we search hard for it, the variety was so big there...
 
And I forgot the Central European I2a2b, present also in possible IE cultures as Unetice and Urnfield, well, I know that there was only a Yamnayan sample which was I2a2a, but it's well stablished that all haplos formed in the steppes and learnt in their schools the academical IE of Pokorny with six or seven h's (the mothers speak an impressive variety of Caucasian languages so that the school was the unique way to learn goodly IE).
 
I've not scientific clues to date (I'm waiting the results of the chiefs of Science competition, people which are pushed to emit quick conclusions almost as quickly as basic poorly knowledged citizens), but good sense tell me that there were differences between several sorts of steppic herders incursions; not too clear, but I think some among the first adventurers were rather males, and they picked foreign local females - as you I don't think the whole IE or IE-icized which settled Europe came without any females, leaving their own females behind them to profit of more attractive western females (and giving these last ones their own children to educate)!
the reality is between: we know the young males in this kinds of societies had sometimes their "ver sacrum", "sacred or holly spring" (?) and they left their well settled tribes, to run adventure and perhaps to look after mercenary business; these young men surely took outside females (Italics did, Celts did, other surely did?), some returned not all of them, other settled in other places, for the most keeping their culture and language ;
SO I bet there have been differences between X-DNA and global auDNA of the pops born after these big pops moves, not so dramatic as said by the first study, but real; all the way, by time, these differences between X and total DNA tend to disappear, evidently.
I add some today old survey I red found (or they believed they had found?) that the global Mediterranean pops from West to East show a tendancy to more level X-DNA than the global DNA of the diverse pops (X less different) -
 
I add it's not always the Steppic males who took farmers female; physical anthropologic surveys about Sredny Stog shew the females were almost 100% Dniestr local pop (HG issued) BUT the males were a mix of local HGs with 'mediterraneanlike' other males, intruders; helas, it didn't say if the tendancy was towards EEF pop or Caucasian or 'east-mediter' types... No Berun, we are not thinking all the unbalanced matings were on the everlasting HG-males / farmers-females (little joke).
 
Cultural affiliation of samples under discussion:

Cultural affiliation of samples analyzed for X chromosome study rebuttal.PNG
 

This thread has been viewed 22205 times.

Back
Top