PDA

View Full Version : Kinship-based social inequality in Bronze Age Europe



Wing Genealogist
10-10-2019, 06:22 PM
https://science.sciencemag.org/content/early/2019/10/09/science.aax6219

I cannot access this paper. Can SKS provide the Y-DNA data for this paper?

Given the fact they have collected 104 remains from Southern Germany, I do expect some U106+ results (which I am looking for) as well as other good data for folks.

K33
10-10-2019, 06:26 PM
It doesn't seem like the DOI is active yet?

Also this thread should probably be moved to the Ancient DNA subforum?

Abstract:
In southern Germany 4000 years ago, several generations of high-ranking families were buried in cemeteries on their farmsteads, with husbands alongside wives, sons, children who died young, and poor, unrelated helpers. But one type of family member was absent from every household plot: adult daughters. Some high-ranking women were present in these Bronze Age graves in the Lech River Valley, south of Augsburg, Germany. But they were all born and reared far from their burial sites. Now, a detailed picture of their social structure has emerged from a remarkable new study. By combining evidence from DNA, artifacts, and chemical clues in teeth, an interdisciplinary team unraveled relationships and inheritance patterns in several generations of high-ranking families buried in cemeteries on their farmstead. In southern Germany 4000 years ago, several generations of high-ranking families were buried in cemeteries on their farmsteads, with husbands alongside wives, sons, children who died young, and poor, unrelated helpers. But one type of family member was absent from every household plot: adult daughters. Some high-ranking women were present in these Bronze Age graves in the Lech River Valley, south of Augsburg, Germany. But they were all born and reared far from their burial sites. Now, a detailed picture of their social structure has emerged from a remarkable new study. By combining evidence from DNA, artifacts, and chemical clues in teeth, an interdisciplinary team unraveled relationships and inheritance patterns in several generations of high-ranking families buried in cemeteries on their farmsteads

Wing Genealogist
10-10-2019, 07:56 PM
As reported in another thread, the supplementary information can be downloaded from: https://science.sciencemag.org/content/suppl/2019/10/09/science.aax6219.DC1

For me, the Bad News is that out of 53 males tested, only 1 was U106+ (HUGO_180sk1) and the test yielded poor results (many SNPs missing, and the SNPs reported almost universally had only a single read).
The Good News is this appears to be the oldest U106 found to date. The paper reported the Median date as 2335.5 calcBCE with a 2 sigma age range 2562–2039 calBCE.

This is reported as a Bell Beaker grave site, so it is also the first confirmed U106 in Bell Beaker.

Chnodomar
10-10-2019, 08:43 PM
As reported in another thread, the supplementary information can be downloaded from: https://science.sciencemag.org/content/suppl/2019/10/09/science.aax6219.DC1

For me, the Bad News is that out of 53 males tested, only 1 was U106+ (HUGO_180sk1) and the test yielded poor results (many SNPs missing, and the SNPs reported almost universally had only a single read).
The Good News is this appears to be the oldest U106 found to date. The paper reported the Median date as 2335.5 calcBCE with a 2 sigma age range 2562–2039 calBCE.

This is reported as a Bell Beaker grave site, so it is also the first confirmed U106 in Bell Beaker.

Where do you even see this information? In Table S1 that sample is merely R1b1a2 and cal BC 2461-2210.

I'd say U106 in southern Germany at that time is pretty unlikely.

Wing Genealogist
10-10-2019, 09:02 PM
Where do you even see this information? In Table S1 that sample is merely R1b1a2 and cal BC 2461-2210.

I'd say U106 in southern Germany at that time is pretty unlikely.

It came from Table S8, which is a separate download from most of the Supplemental Information (it downloads a zipped file containing several spreadsheet tables). They used the M405 SNP name (row 1638) and HUGO_180sk1 is in column Z.

Wing Genealogist
10-10-2019, 09:08 PM
From the supplemental information:

SI 4 – Y-chromosomal haplogroup assignment
Assignment for Y-chromosomal haplogroups was performed following the nomenclature of the
International Society of Genetic Genealogy (ISOGG) version 11.110 (retrieved 21 April 2016,
http://www.isogg.org). Haplogroup defining SNPs that were covered in any of the 53 males are
listed in table S8. Calls on C-to-T or G-to-A SNPs were only considered reliable when more than
50% of calls showed the allele (table S8).
The finding of a male sex bias in steppe-related admixture is strengthened by the distribution of
Y-chromosomal haplogroups that is dominated by the R1b-lineage that entered western Europe
in the third millennium BCE in steppe-related populations.
Of the 40 men from the Lech valley that had reliable derived SNP calls on positions informative
about Y-chromosomal haplogroups, 17 could be assigned to R1b-P312/S116* (R1b1a2a1a2*),
with no more derived alleles further downstream called. This lineage was shown to be prevalent
among Beaker-complex-associated males outside of Iberia, and it has been suggested that these
people played a major role in its distribution across central and western Europe (16). Ten
additional men had derived alleles placing them on the lineage spanning P1-M45 to R1bL11/P310 (R1b1a2a1a), but were not covered at P312/S116.
On the other hand, one individual associated with the Corded Ware complex (ALT_4) could be
assigned to R1b-L11/P310 (R1b1a2a1a) and did not have the derived downstream allele defining
P312/S116.
Only three individuals from the Lech valley fell into clades distinct from P1-M45 and its
subclades:
• HUGO_169Sk1 and UNTA58_68Sk1, both dated to the BBC, fall on the Y clade G2,
which was among the most frequent in Europe in the preceding Neolithic period (10, 115, 118,
132), and was likely carried there by the expansion of early farmers out of Anatolia (12, 110,
116).
• OBKR_117, dated to the EBA, could be assigned to the I clade, already present in postglacial Europe (1, 35, 117). Interestingly, this male does not have any relatives at the same burial
site, where the three other males for whom assignments could be made all carry P312/S116.

Wing Genealogist
10-10-2019, 09:15 PM
While the above synopsis of their Y-DNA analysis states they use the ISOGG 2016 tree, luckily they do include the SNP name as well as the ISOGG "long-hand" name. This makes things MUCH easier. The only minor quibble is that sometimes the "lab name" they use is not the name we are used to. For example, they use M405 for the U106 SNP.

Principe
10-10-2019, 09:38 PM
This is quite huge news!! Looks like CWC was the ancestor of R1b-L151 and its downstreams, now all that leaves is where L51 is from and L151's brother clade Z2118 honestly looking at Yfull seems to be somewhere around the area. Of course PF7562 is still missing from ancient remains.

In my opinion this also leaves a big questions for Yamnaya? If CWC is ancestral to Indo-Iranian-Aryan/Balto-Slav/Celto-Italic/Germanic what was Yamnaya ancestral too? Is CWC an offshoot, if someone can answer me this! Or will Yamnaya represent the ancestral branches to Balkan and Armenian language family, or perhaps Yamnaya is really Proto-Anatolian?

I guess this would make sense in a way that BBC would be ancestral to the Western and Central European I-E branches and Sintasha being ancestral to Satem languages. Again maybe Yamnaya itself could explain other branches like Hellenic, Albanian, Armenian and Phrygian, which perhaps explains why they are so hard to place in the tree? I guess when they find PF7562 in ancient dna will we know where older branches of PIE split? I guess the ancestor of Yamnaya will be PIE, my honest vote would go to Dnieper–Donets culture.

razyn
10-10-2019, 09:59 PM
I don't quite agree with the statement
Of the 40 men from the Lech valley that had reliable derived SNP calls on positions informative
about Y-chromosomal haplogroups, 17 could be assigned to R1b-P312/S116* (R1b1a2a1a2*),
with no more derived alleles further downstream called.

I got the 17 P312+ samples, but two of them (cols. R and BB ) did in fact have downstream calls for U152+. They were looking for 11 subclades of DF27, especially several that might be expected in a Basque population, but didn't consider any of the scattered results (rows 1686-1696) to be conclusive. And I think that's probably right, but doesn't in itself mean that none of these fifteen P312* guys were DF27+. As we are accustomed to finding, they seem not to have tested for that.

Btw I don't think they found L21+, either.

Edit: I note that the U152 mutation is C-T, and they disregarded C-T because it could be an artifact of post-mortem deterioration. OTOH it also could be real.

MitchellSince1893
10-11-2019, 12:43 AM
...The Good News is this appears to be the oldest U106 found to date. The paper reported the Median date as 2335.5 calcBCE with a 2 sigma age range 2562–2039 calBCE.

This is reported as a Bell Beaker grave site, so it is also the first confirmed U106 in Bell Beaker.
Congrats!
So the oldest U106 to date is Bell Beaker? Well, that is a game changer.

Wing Genealogist
10-11-2019, 01:08 AM
Congrats!
So the oldest U106 to date is Bell Beaker? Well, that is a game changer.

Not necessarily a game changer. Given the vast number of ancient Bell Beaker already published, it is clear U106 did not play a significant role in the Bell Beaker Culture.

MitchellSince1893
10-11-2019, 01:31 AM
Not necessarily a game changer. Given the vast number of ancient Bell Beaker already published, it is clear U106 did not play a significant role in the Bell Beaker Culture.

IMO, It's a big deal. Until today, many argued that U106 had to originate in Battle Axe as the previously oldest U106 sample was from Battle Axe/Sweden site. So how is an older U106 sample found in a different culture not a significant discovery?

The vast majority of Corded Ware samples have been R1a; and until today, it has been successfully argued that L51 and descendants weren't part Corded Ware. The discovery of L151 in CW...even if it's a contemporary of the oldest P312 means that argument is no longer valid.

I've heard that same line of argument used for U106 not being part of Bell Beaker. Today's discovery of the oldest U106 in Bell Beaker shows that U106 was part of BB...even if it was a small part...just like L151/L51 may have been a small part of Corded Ware.

Radboud
10-11-2019, 06:45 AM
As reported in another thread, the supplementary information can be downloaded from: https://science.sciencemag.org/content/suppl/2019/10/09/science.aax6219.DC1

For me, the Bad News is that out of 53 males tested, only 1 was U106+ (HUGO_180sk1) and the test yielded poor results (many SNPs missing, and the SNPs reported almost universally had only a single read).
The Good News is this appears to be the oldest U106 found to date. The paper reported the Median date as 2335.5 calcBCE with a 2 sigma age range 2562–2039 calBCE.

This is reported as a Bell Beaker grave site, so it is also the first confirmed U106 in Bell Beaker.


I am not sure if this result is legit, because it's based on a single read, it's a low quality sample and many SNP's are missing like you said.

Last year, there were two similar cases: two Bell Beakers, a Hungarian and a Czech one, were thought to be known as Z18 and Z9 respectively. However, these findings were not confirmed by Williamson. Williamson suggested that it was a result of deamination, which happens for aDNA, and results in spurious C->T (and G->A in results) mutations.

Apparently, the authors of this paper were not confident enough to classify HUGO_180sk1 as U106/M405, but rather as just R1b1a2. In this case, the best thing to do is to let an expert like Willamson take a look at this sample to confirm or deny/doubt that he is U106 IMO. Are there any BAM files available?

Wing Genealogist
10-11-2019, 08:25 AM
I am not sure if this result is legit, because it's based on a single read, it's a low quality sample and many SNP's are missing like you said.

Last year, there were two similar cases: two Bell Beakers, a Hungarian and a Czech one, were thought to be known as Z18 and Z9 respectively. However, these findings were not confirmed by Williamson. Williamson suggested that it was a result of deamination, which happens for aDNA, and results in spurious C->T (and G->A in results) mutations.

Apparently, the authors of this paper were not confident enough to classfiy HUGO_180sk1 as U106/M405, but rather as just R1b1a2. In this case, the best thing to do is to let an expert like Willamson take a look at this sample to confirm or deny/doubt that he is U106 IMO. Are there any BAM files available?

You are correct in this caution, as the U106 mutation (C->T) is subject to deamination, so a single read is anything but conclusive. Additionally, there are no reads for the vast majority of clades below M269, so we cannot rule out this individual being P312+ or some other clade below M269. We do really need a close examination of the BAM file by an expert.

Being human, it's hard not to get excited when a somewhat surprising result (such as U106 in Bell Beaker) shows up. Hopefully the BAM file will show up results for SNPs not documented in table S8 and we are able to more confidently place this sample.

EDIT: It appears the researchers have discovered a way to differentiate between a true thymine call and a false call due to the deamination process. Thus the single read for U106 was likely not due to deamination. Even so, a single call is still not enough evidence to state with ANY confidence the call is legitimate. There are a number of other reasons where a single call is in error.

Ruderico
10-11-2019, 08:31 AM
IMO, It's a big deal. Until today, many argued that U106 had to originate in Battle Axe as the previously oldest U106 sample was from Battle Axe/Sweden site. So how is an older U106 sample found in a different culture not a significant discovery?

The vast majority of Corded Ware samples have been R1a; and until today, it has been successfully argued that L51 and descendants weren't part Corded Ware. The discovery of L151 in CW...even if it's a contemporary of the oldest P312 means that argument is no longer valid.

I've heard that same line of argument used for U106 not being part of Bell Beaker. Today's discovery of the oldest U106 in Bell Beaker shows that U106 was part of BB...even if it was a small part...just like L151/L51 may have been a small part of Corded Ware.

I agree, it opens up all kinds of possibilities, even if it was a minor clade amongst them. Assuming, that is, that it's not an extinct subclade of U106

Generalissimo
10-11-2019, 08:58 AM
The authors were very careful with the Y-hg classifications in this paper, and did a good job, unlike what we've seen in other papers.

So if the results don't say U106, then it's unlikely to be U106.

nuadha
10-11-2019, 01:53 PM
Am I reading Table S9 correctly? It looks like Tauber CWC is just above 1/3 steppe while Lech BB is around 1/2 steppe.

Then, EBA and MBA Lech tends to be even higher in steppe.

Chnodomar
10-11-2019, 05:28 PM
Am I reading Table S9 correctly? It looks like Tauber CWC is just above 1/3 steppe while Lech BB is around 1/2 steppe.

Then, EBA and MBA Lech tends to be even higher in steppe.

Good catch, I'd say they simply entirely transposed the columns in that table:

TauberCWC:
Globular_Amphora 0.68
Yamnaya 0.32

The other way round that would make perfect sense and would be a typical CWC result and would fit figure S8:

33820

"Fig. S8.f4-statistics showing differences in ancestry in populations grouped by period. (A) An increase in affinity to ancestry related to Anatolia Neolithic over time. Males and females grouped together shown as upward and downward pointing triangles, respectively.(B) and (C) f4-statistics of the form f4(Mbuti, Test; earlier Lech population, later Lech population) become significantly positive when Test is a Neolithic or Chalcolithic population, indicating progressive admixture from populations with comparatively more Neolithic-related ancestry."

Table S9 certainly disagrees with any other table or text in that PDF:

"Modeling with qpAdm the period groupings as three way mixtures of WHG, Anatolia_Neolithic and Yamnaya_Samara using the parameters allsnps: YES, we observe an increase over time in the proportion of Middle Neolithic ancestry in the southern German population and correspondingly, a slight decrease in steppe ancestry over time (Fig. 2, table S7).

To examine if we could more accurately identify the Middle Neolithic genetic stratum involved in the admixture event that formed the Lech populations we explicitly tested two-way admixture models with qpAdm, using Yamnaya” from Samara as first source and different European Middle Neolithic and Chalcolithic populations as second source. The best fitting models were obtained with the farmers of the Globular Amphora culture, the Middle Neolithic population of France and Chalcolithic Iberia, who all gave p-values over 0.05 for the Lech BBC, EBA and MBA populations when used as a source (table S9)."

And not even the description of table S9 is correct, it's obviously not a three-way admixture model and the above text that references it clearly states it's a two-way model.

They have some very obvious technical mistakes in the Supplementary Information, unfortunately. Just look at Fig. S5, too. The sample and culture names are entirely unreadable.

K33
10-11-2019, 08:23 PM
Everyone including myself is now toying with the idea that Yamnaya wasn't directly involved in the Indo-Europeanization of Europe, but the exact process is elusive. I've tried to do some nMonte modelling of all the extant CWC and Bell Beaker samples using Eneolithic steppe (ie, non-Yamnaya/pre-Yamnaya) proxies. But CWC/BBC requires BOTH:

a) a "North Pontic" source like Sredny Stog (Sredny Stog_En_II is I believe from Alexandria, Ukraine; alternatively the Dereivka_En_1/2 samples are nearby), AND
b) an "East Pontic"/Caucasus steppe source like Progress/Vonyuchka

Now Progress/Vonyuchka is basically contemporary with the Alexandria sample (~4000-4200 BCE). The Dereivka_En_I and Dereivka_II samples are dated 3200 BCE, temporally closer to CWC, but these produce worse fits with the CWC/BBC samples. They look basically like WHG-shifted versions of the Alexandria sample.

Using distal models, Progress is 2/3 Samara HG and 1/3 CHG; Sredny_II (Alexandria) is more like 2/3 Samara and 1/3 ENF. Removing either Progress OR Sredny Stog severely damages all the fits for CWC/BBC samples. I think David has speculated on this before, but the Repin Culture seems geographically midway between Sredny Stog and Progress.

Current theory:
- One Repin group carrying R1a-M417 rapidly worked northwest thru the forest-steppe circa 3500 BCE, while the cousins they left behind were folded into Yamnaya on the steppe proper
- And the big surprise: this splinter Repin group bumped into the remnants of the Dnepier-Donets hunters somewhere near the Ukraine-Belarus border; DD guys contributed R-L51 to the mix

(bold rows below indicate poor fits at >3.5% distance)

https://i.imgur.com/MLMIhhu.jpg

falconson1
10-11-2019, 09:12 PM
Alas, the Y haplogroup data as depicted in Supplementary Material Table S8, with the Y haplogroups presented both as ISOGG letters and numbers, and as supposed terminal SNPs, has a vast number of "holes". There is really little to be learned until a true expert (e.g., "Williamson") "reads" the associated BAM files and confirms what is and is not correct. Just looking at R-U152 related SNPs, U152 is "greyed out" (not working up to defined standard), although two individuals in the sample group were a "hit" on one run. Oddly the most common downstream SNP, L2, is not even included - although downstream Z367 was ("worked" but no individuals positive for this SNP - and further downstream the relatively common L20 is grayed out). The handfull of other U152 - related SNPs include many I don't recognize, although L4 worked well (no derived samples), but as I recall is confined to the Ashkenazi population and not likely to be "ancient". I am beginning to doubt that this paper will end up being in the same league as the other major Bell Beaker study, Olalde et al., 2018. Perhaps the soil conditions were not as conducive to being able to avoid deamination issues (e.g., more acidic?). Hopefully as I type this, someone is assessing the BAM file and we will soon have more clarity. It was interesting to note that another Bell Beaker candidate SNP, L21 (predominant today in for example Ireland), worked relatively well, but zero samples appear to have been derived for this SNP.

Dieu
10-11-2019, 09:20 PM
What was the Y-haplo and autosomale profile of high status and low status individuals ?

Ryukendo
10-11-2019, 09:36 PM
The authors ran a f4 comparison in autosomal ancestry between males in graves with relatives, vs males without relatives. (Males with relatives all had R1b, while males without had R1b and other Y chromosomes. All males belonged either to the familial lineage within their respective graves, i.e. "group A", in which case their mothers, wives etc were sometimes present along with other relatives along the male line, or to non-familial groups in which case no relatives were present, in "group B".)

Their results:


Results are listed in table S12. We could detect no significant
dissymmetry in shared drift between the groups, as Z-scores that fell above |3| were expected due
to multiple testing (fig. S29).

There were no differences in phenotypic alleles either. The differences in R1b representation in high-status vs low-status groups is maybe not so robust, as there were only 3 non-R1b Y chromosomes in the entire sample.

Dieu
10-11-2019, 10:57 PM
Current theory:
- One Repin group carrying R1a-M417 rapidly worked northwest thru the forest-steppe circa 3500 BCE, while the cousins they left behind were folded into Yamnaya on the steppe proper
- And the big surprise: this splinter Repin group bumped into the remnants of the Dnepier-Donets hunters somewhere near the Ukraine-Belarus border; DD guys contributed R-L51 to the mix



What about the R1b Z2103 in your scenario ? Z2103 in Yamna and L51 in Dnieper Donets doesn't make much sense unless L23 is dnieper donets haplogroup or even samara in origin.

rms2
10-11-2019, 11:31 PM
IMO, It's a big deal. Until today, many argued that U106 had to originate in Battle Axe as the previously oldest U106 sample was from Battle Axe/Sweden site. So how is an older U106 sample found in a different culture not a significant discovery?

The vast majority of Corded Ware samples have been R1a; and until today, it has been successfully argued that L51 and descendants weren't part Corded Ware. The discovery of L151 in CW...even if it's a contemporary of the oldest P312 means that argument is no longer valid.

I've heard that same line of argument used for U106 not being part of Bell Beaker. Today's discovery of the oldest U106 in Bell Beaker shows that U106 was part of BB...even if it was a small part...just like L151/L51 may have been a small part of Corded Ware.

Except we're not really sure that BB guy is derived for U106.

I think U106 may have been part of BB, but probably in BB's Northern Province, which has not been explored yet.

dsm
10-12-2019, 12:00 AM
Except we're not really sure that BB guy is derived for U106.

I think U106 may have been part of BB, but probably in BB's Northern Province, which has not been explored yet.

I am with rms2 on this. While 'betting' is risky, I am inclined to bet that there isn't a U106 in that DNA. But, lets hope Alex offers his expert opinion soon.

rms2
10-12-2019, 12:38 AM
I wish this hobby moved a little faster.

I have to find a faster moving hobby, like watching the first maple sap run in spring, or watching paint dry.

Pop some popcorn, you know.

Silesian
10-12-2019, 02:40 AM
Everyone including myself is now toying with the idea that Yamnaya wasn't directly involved in the Indo-Europeanization of Europe, but the exact process is elusive. I've tried to do some nMonte modelling of all the extant CWC and Bell Beaker samples using Eneolithic steppe (ie, non-Yamnaya/pre-Yamnaya) proxies. But CWC/BBC requires BOTH:

a) a "North Pontic" source like Sredny Stog (Sredny Stog_En_II is I believe from Alexandria, Ukraine; alternatively the Dereivka_En_1/2 samples are nearby), AND
b) an "East Pontic"/Caucasus steppe source like Progress/Vonyuchka

Now Progress/Vonyuchka is basically contemporary with the Alexandria sample (~4000-4200 BCE). The Dereivka_En_I and Dereivka_II samples are dated 3200 BCE, temporally closer to CWC, but these produce worse fits with the CWC/BBC samples. They look basically like WHG-shifted versions of the Alexandria sample.

Using distal models, Progress is 2/3 Samara HG and 1/3 CHG; Sredny_II (Alexandria) is more like 2/3 Samara and 1/3 ENF. Removing either Progress OR Sredny Stog severely damages all the fits for CWC/BBC samples. I think David has speculated on this before, but the Repin Culture seems geographically midway between Sredny Stog and Progress.

Current theory:
- One Repin group carrying R1a-M417 rapidly worked northwest thru the forest-steppe circa 3500 BCE, while the cousins they left behind were folded into Yamnaya on the steppe proper
- And the big surprise: this splinter Repin group bumped into the remnants of the Dnepier-Donets hunters somewhere near the Ukraine-Belarus border; DD guys contributed R-L51 to the mix

(bold rows below indicate poor fits at >3.5% distance)

https://i.imgur.com/MLMIhhu.jpg

I'm toying with the idea that the Bell Beaker sample with Narva is related to the same branch of R1b within [Yamnaya burials] same burial mounds of Progress R1b-V1636+

Beaker_Hungary_I2787
Beaker_Bavaria 0.442±0.045
Yamnaya_Samara 0.558±0.045
chisq 8.562
tail prob 0.73982
Full output

Beaker_Hungary_I2787
Beaker_Czech 0.441±0.045
Yamnaya_Samara 0.559±0.045
chisq 10.009
tail prob 0.61513
Full output

Beaker_Hungary_I2787
Beaker_The_Netherlands 0.576±0.062
Yamnaya_Samara 0.424±0.062
chisq 11.469
tail prob 0.489238
Full output
http://images.devs-on.net/Image/eyQR1dVGP2h59E4Y-Region.png

Radboud
10-12-2019, 02:45 PM
You are correct in this caution, as the U106 mutation (C->T) is subject to deamination, so a single read is anything but conclusive. Additionally, there are no reads for the vast majority of clades below M269, so we cannot rule out this individual being P312+ or some other clade below M269. We do really need a close examination of the BAM file by an expert.

Being human, it's hard not to get excited when a somewhat surprising result (such as U106 in Bell Beaker) shows up. Hopefully the BAM file will show up results for SNPs not documented in table S8 and we are able to more confidently place this sample.

EDIT: It appears the researchers have discovered a way to differentiate between a true thymine call and a false call due to the deamination process. Thus the single read for U106 was likely not due to deamination. Even so, a single call is still not enough evidence to state with ANY confidence the call is legitimate. There are a number of other reasons where a single call is in error.

The BAM files are available here (https://www.ebi.ac.uk/ena/data/view/PRJEB34400). I have sent a Williamson a message.

Principe
10-12-2019, 05:07 PM
ALT_4, is R1b-L52*

https://www.yfull.com/tree/R-L52/

I think the debate is over!

Seeing that ALT_4 is L52* and not even L151 pretty confirms that L151 arose in CWC and thus so did its branches.

ArmandoR1b
10-12-2019, 05:54 PM
ALT_4, is R1b-L52*

https://www.yfull.com/tree/R-L52/

I think the debate is over!

Seeing that ALT_4 is L52* and not even L151 pretty confirms that L151 arose in CWC and thus so did its branches.

ALT_4 is not known to be L52*. Where do get that crazy idea? It is not known if positive or negative for L52/PF6541, P311/PF6545/S128, PF6540/YSC0000082, or CTS7650/FGC44/PF6544. It is also not known if positive or negative for L151/PF6542, L11, PF6543/S1159/YSC0000191. There is no coverage of those SNPs in the BAM files or in Table8 of the study.

ArmandoR1b
10-12-2019, 06:00 PM
R.Rocca provided a link to the BAM files earlier in a different thread. They are at https://www.ebi.ac.uk/ena/data/view/PRJEB34400

Principe
10-12-2019, 06:28 PM
ALT_4 is not known to be L52*. Where do get that crazy idea? It is not known if positive or negative for L52/PF6541, P311/PF6545/S128, PF6540/YSC0000082, or CTS7650/FGC44/PF6544. It is also not known if positive or negative for L151/PF6542, L11, PF6543/S1159/YSC0000191. There is no coverage of those SNPs in the BAM files or in Table8 of the study.

He got a call for P310, all the other snps are no calls.

ArmandoR1b
10-12-2019, 07:10 PM
He got a call for P310, all the other snps are no calls.

So what you are saying is that you do not understand what it means that he got a call for P310, all the other snps are no calls. That means that he can't be called L52* due to the fact that there are no calls on L52/PF6541, P311/PF6545/S128, PF6540/YSC0000082, or CTS7650/FGC44/PF6544 and L151/PF6542, L11, PF6543/S1159/YSC0000191 due to no coverage.

Principe
10-12-2019, 07:43 PM
So what you are saying is that you do not understand what it means that he got a call for P310, all the other snps are no calls. That means that he can't be called L52* due to the fact that there are no calls on L52/PF6541, P311/PF6545/S128, PF6540/YSC0000082, or CTS7650/FGC44/PF6544 and L151/PF6542, L11, PF6543/S1159/YSC0000191 due to no coverage.

He can also form his own clade would P310* be better to say?

The major point would be we have a pre L151 in a CWC sample which is a significant finding.

ArmandoR1b
10-12-2019, 07:59 PM
He can also form his own clade would P310* be better to say?

No, the * after the SNP means that all downstream SNPs are negative. That can't be said of anyone that has no-calls of phylogenetic equivalents and of immediate downstream SNPs. A correct way to say it is - P310+ and unknown for phylogenetic equivalents and of immediate downstream SNPs.

Another, but longer, way to say it is include all phylogenetic equivalents with ? after the unknown as follows P310+, L52/PF6541?, P311/PF6545/S128?, PF6540/YSC0000082?, CTS7650/FGC44/PF6544? and L151/PF6542?, L11?, PF6543/S1159/YSC000019?

We have no idea if he can form his own clade because the coverage is too low. So it is unknown if he can form his own clade. A lack of evidence is not evidence.

We need older specimens from further east that test positive for P310 or a phylogenetic equivalent to solidify that the ancestors lived in that region at that time. Until then, I repeat, a lack of evidence is not evidence.

ArmandoR1b
10-12-2019, 08:02 PM
The major point would be we have a pre L151 in a CWC sample which is a significant finding.

We don't know if we have a pre L151 in a CWC sample because ALT_4 has no-calls for L11/S127/PF6539, YSC0000191/PF6543/S1159, and L151/PF6542. A lack of evidence is not evidence.

Wing Genealogist
10-12-2019, 09:14 PM
I noted where virtually all of the samples have two different results, one
Model: NextSeq 500 which has a relatively low number of SNPs And
Model: Illumina HiSeq 4000 with MANY more SNPs.

Is it possible for the NextSeq 500 results to have reads for SNPs not read by the HiSeq 4000, or is it fairly safe to basically ignore the NextSeq 500 results.

artemv
10-12-2019, 09:55 PM
Everyone including myself is now toying with the idea that Yamnaya wasn't directly involved in the Indo-Europeanization of Europe, but the exact process is elusive. I've tried to do some nMonte modelling of all the extant CWC and Bell Beaker samples using Eneolithic steppe (ie, non-Yamnaya/pre-Yamnaya) proxies. But CWC/BBC requires BOTH:

a) a "North Pontic" source like Sredny Stog (Sredny Stog_En_II is I believe from Alexandria, Ukraine; alternatively the Dereivka_En_1/2 samples are nearby), AND
b) an "East Pontic"/Caucasus steppe source like Progress/Vonyuchka

Now Progress/Vonyuchka is basically contemporary with the Alexandria sample (~4000-4200 BCE). The Dereivka_En_I and Dereivka_II samples are dated 3200 BCE, temporally closer to CWC, but these produce worse fits with the CWC/BBC samples. They look basically like WHG-shifted versions of the Alexandria sample.

Using distal models, Progress is 2/3 Samara HG and 1/3 CHG; Sredny_II (Alexandria) is more like 2/3 Samara and 1/3 ENF. Removing either Progress OR Sredny Stog severely damages all the fits for CWC/BBC samples. I think David has speculated on this before, but the Repin Culture seems geographically midway between Sredny Stog and Progress.

Current theory:
- One Repin group carrying R1a-M417 rapidly worked northwest thru the forest-steppe circa 3500 BCE, while the cousins they left behind were folded into Yamnaya on the steppe proper
- And the big surprise: this splinter Repin group bumped into the remnants of the Dnepier-Donets hunters somewhere near the Ukraine-Belarus border; DD guys contributed R-L51 to the mix

(bold rows below indicate poor fits at >3.5% distance)

https://i.imgur.com/MLMIhhu.jpg

Thank you.
Sredny Stog / Progress ratio differs in a very wide range, we can only suggest that several groups with steppe ancestry, that contributed to CWC, where not homogenous and included groups with slightly different ancestry.
Anyway, Sredny Stog and Progress are too old cultures, there must be a more recent proxy, we have to wait for their DNA.
At the moment I think the problem is that we have no aDNA from forest-steppe or forest zones of East Europe (second half of 4th millenium BCE) - territory of modern Belarus/North Ukarine/Central Russia (somewhere in Mosow-Bryank-Smolensk triangle). I guess there is a good chance they contributed to CWC genetically.

What looks quite evident but still important - is the role GAC played in emerging of CWC and their genetic contribution to the modern Europeans. This agrees with Mittnik results, and doesn't contradict what we know from recent Malmstrom paper (about BAC ancetry). Looks like significant part of EEF ancestry in modern North Europeans comes from GAC and was brought afterwards to all other parts of Europe, settled by CWC/BBC.

rms2
10-13-2019, 12:57 PM
We don't know if we have a pre L151 in a CWC sample because ALT_4 has no-calls for L11/S127/PF6539, YSC0000191/PF6543/S1159, and L151/PF6542. A lack of evidence is not evidence.

True, but having a P310+ call is pretty significant in and of itself.

Wish the dates on ALT_4 were a little older though, and I wish they had tested more Corded Ware remains.

If Kurgan Bell Beaker did emerge as a sect within Corded Ware, it certainly looks like the Corded Ware of the Lech Valley was predominantly if not exclusively R1b-P310, most of it P312 or on the line leading to P312 (yes, I realize ALT_4 was negative for P312).

ArmandoR1b
10-13-2019, 05:30 PM
True, but having a P310+ call is pretty significant in and of itself.
Yes it is but it is still important to be accurate.


Wish the dates on ALT_4 were a little older though, and I wish they had tested more Corded Ware remains.
Yes, we need much older remains that are L151+P312-U106-, L52+L151-, and L51+L52- and each of those to be significantly older than the other so we can see the trail. It will be really exciting if some of the phylogenetic equivalents can be broken up too by ancient remains such as L151+L11- or L11+L151- or something similar upstream.


If Kurgan Bell Beaker did emerge as a sect within Corded Ware, it certainly looks like the Corded Ware of the Lech Valley was predominantly if not exclusively R1b-P310, most of it P312 or on the line leading to P312 (yes, I realize ALT_4 was negative for P312).

I don't see a read for P312 (position 22157311) in the BAM files of ALT_4. I'm not sure what is going on there. I know that Table8 shows 0.0% (2) for P312 so I'm not sure if it was an error or if the coverage of P312 (position 22157311) was removed prior to the BAM file being uploaded.

alan
10-14-2019, 12:03 PM
One factor that may produce strange apparent family style groupings who are not biological relatives (other than exogenous marriage) is the tradition noted in Celtic societies of families fostering their sons from later childhood until they reach adulthood (then return) with another house, clan or elite family in a other territory. Presumably this originated as another way of cementing alliances and gaining knowledge and better understanding of the politics of other tribes.

dsm
10-15-2019, 08:48 AM
ALT_4, is R1b-L52*

https://www.yfull.com/tree/R-L52/

I think the debate is over!

Seeing that ALT_4 is L52* and not even L151 pretty confirms that L151 arose in CWC and thus so did its branches.

That info YFull tree) might be useful if the YFull tree was accurate *but it isn't* therefore is a dubious source for confirming anything related to L151. Just look at their TMRCA for CTS4528 = 4800, then go up one level to S1194 (YFull don't even list S1194) but it is by their default = 4800. Then go to L151 and yet again TMRCA is 4800. It us a mess. Using their own average requires adding 130 for each SNP above CTS4528.

To their credit YFull have advised me that they agree they need to correct their L151 tree. When that has happened, I may trust you using them as an authority.

Silesian
10-16-2019, 11:21 PM
Everyone including myself is now toying with the idea that Yamnaya wasn't directly involved in the Indo-Europeanization of Europe, but the exact process is elusive. I've tried to do some nMonte modelling of all the extant CWC and Bell Beaker samples using Eneolithic steppe (ie, non-Yamnaya/pre-Yamnaya) proxies. But CWC/BBC requires BOTH:



Here is another genetic model using different method [IMO] to fit better with Progress and or Yamnaya for both Baltic Corded and Bell Beaker.
Yamnaya being the basal clade,of the now extinct steppe [Yersinia pestis] that was tracked by mutations to the following pool of R1b/R1a. You can draw your own conclusions how Corded Ware sandwiched in the middle of two R1b populations acquired the bacteria, however keep in mind, Lithuanian R1a had a typical hammer-headed style bone pin[very popular with Yamnaya]

Same branch, steppe Yersinia pestis. chronological mutation order
>Yamnaya-Caucasus RK1001[Z2109?]>Afanasievo Rise 509[z2109?]> Corded Ware Gyvakarai1[R1a1a1b (Z645)] Vucedol I3499[Z2109?]> Corded Ware Kunila2 [R1a1a1b (Z645)]>Bell Beaker E09569[U152/S28]>6Post?>Rise 505 [Z2109?]>Armenia RISE397 [Z2106+]

http://images.devs-on.net/Image/IHUqd06eYNTVM9pQ-Region.png


https://media.nature.com/original/nature-assets/nature/journal/v555/n7695/extref/nature25738-s2.pdf

Wing Genealogist
10-17-2019, 11:48 PM
I have asked Williamson to analyze the BAM files for HUGO_180sk1. There were two BAM files for each sample (one run on an illumina NextSeq 500 series machine and the other on an illumina HiSeq4000 series machine). Unfortunately BOTH files show very poor SNP calling. He was unable to discover ANY positive SNP results below R-M269.

I believe he has automated his BAM file analysis results to a degree, and likely eliminates any single read C->T calls due to the potential for deamination. This study has already used other techniques to reduce/eliminate these errant calls. As stated earlier, the single positive read for M405 (U106) is very speculative and can easily be an error of one sort or another. Therefor, this sample cannot be called as positive for anything below R-M269.

rms2
10-18-2019, 09:30 PM
. . .
I don't see a read for P312 (position 22157311) in the BAM files of ALT_4. I'm not sure what is going on there. I know that Table8 shows 0.0% (2) for P312 so I'm not sure if it was an error or if the coverage of P312 (position 22157311) was removed prior to the BAM file being uploaded.

I wrote that he was negative for P312 based on the following from page 25 of the Mittnik et al Supplementary Material.



On the other hand, one individual associated with the Corded Ware complex (ALT_4) could be assigned to R1b-L11/P310 (R1b1a2a1a) and did not have the derived downstream allele defining P312/S116.


That seems an odd way to say there was no call at P312. If there was simply no call at P312, they would not be able to confidently say ALT_4 "did not have the derived downstream allele defining P312/S116." They could only say they did not know one way or the other.

rms2
10-18-2019, 09:41 PM
I'm thinking the fact that ALT_4 was R1b-P310, very typically Corded Ware in having a lot of steppe dna, and that Kurgan Bell Beaker appears to have been derived from Corded Ware (there are too many similarities), that means that eventually more L51 (of whatever subclade) will be found in Corded Ware.

If you all have not read Corded Ware from East to West (https://www.encyclopedia.com/humanities/encyclopedias-almanacs-transcripts-and-maps/corded-ware-east-west), it's worth your time to do so.

It's an informative article.

Now that Mittnik et al has put paid to the goofy idea that Beaker got its steppe dna from Corded Ware women, it's time to think more seriously about the genesis of Kurgan Bell Beaker. Since its autosomal profile does not seem to be a cross between Yamnaya and Vucedol, as Gimbutas suggested, but is closer to Corded Ware, and since the Neolithic farmer component in Kurgan Bell Beaker appears to have been contributed by GAC and/or TRB, pointing to a track north of the Carpathians, it's time to consider how Kurgan Bell Beaker might have developed from Corded Ware and just where that first happened.

rms2
10-19-2019, 11:00 AM
Probably everyone has seen this video, but if you haven't, it's worth watching the whole thing, and if you have seen it, it's worth watching again. IMHO, it's the best of the crop of recent archaeogenetics videos.


https://www.youtube.com/watch?v=oi1C1XMYU2Q&fbclid=IwAR0nYBwQxNeI53TF_qeZRZB8g5otVbLT2d19tLFNq TQpnQkNSC5-UX2txg4

razyn
11-23-2019, 05:51 AM
Speaking of well worth watching, the lead author of this paper just did a slide talk about it that has been linked here (by Rosenfeld, IIRC). https://www.youtube.com/watch?v=syej3SWtRlI It runs a bit over an hour, with some old guys being pompous at the end, but most of it is quite interesting. I clipped out a screen shot of the very personable Dr. Mittnik:
34845

ArmandoR1b
11-23-2019, 03:27 PM
I wrote that he was negative for P312 based on the following from page 25 of the Mittnik et al Supplementary Material.

That seems an odd way to say there was no call at P312. If there was simply no call at P312, they would not be able to confidently say ALT_4 "did not have the derived downstream allele defining P312/S116." They could only say they did not know one way or the other.

I just saw this response. I don't always get a notification of quotes. I wrote the author to find out why the BAM file does not include P312 yet Table8 shows 0.0% (2) for P312 but I never got a response. I have never had a problem finding coverage of an SNP, or it's phylogenetic equivalent, in a BAM file that is reported in an academic study. That is the only one that does not have coverage. I strongly feel that if R.Rocca or one of the other people that also know how to look at BAM files would also fail to find coverage of P312. The reason there is no coverage should be explained so we can have the correct information.

edit: I just emailed two of the other authors

Arame
11-26-2019, 10:33 AM
HUGO_169Sk1 is marked as G2 from a Bell Beaker culture. It would be interesting to see wether it is positive for P303. An SNP that is found today in Alpes.

ArmandoR1b
11-28-2019, 12:59 AM
HUGO_169Sk1 is marked as G2 from a Bell Beaker culture. It would be interesting to see wether it is positive for P303. An SNP that is found today in Alpes.

Both the BAM file and TableS8 show him to be positive for M3267 and M3483. There is no coverage of P303 or any of the subclades found in the ISOGG tree. The file size is small meaning that there is very little coverage of the genome of HUGO_169Sk1.

S04ancap
12-30-2019, 08:17 PM
Both the BAM file and TableS8 show him to be positive for M3267 and M3483. There is no coverage of P303 or any of the subclades found in the ISOGG tree. The file size is small meaning that there is very little coverage of the genome of HUGO_169Sk1.

It looks like he may be G-L497 which would be an awesome find given the date of 4600 ybp and only the second BC L497 sample we have (the other being the Z1815 Baden sample from Hungary).

Chnodomar
12-30-2019, 09:27 PM
It looks like he may be G-L497 which would be an awesome find given the date of 4600 ybp and only the second BC L497 sample we have (the other being the Z1815 Baden sample from Hungary).

There seem to be more, actually. At least Neolithic samples from the Vinca and Cucuteni-Trypillian cultures and a 800 BC Hallstatt culture sample: https://anthrogenica.com/showthread.php?9495-Origins-of-haplogroup-G-L497

palamede
12-30-2019, 09:53 PM
It looks like he may be G-L497 which would be an awesome find given the date of 4600 ybp and only the second BC L497 sample we have (the other being the Z1815 Baden sample from Hungary).

In the study of ancient DNA "The Genomic History Of Southeastern Europe", there is a G2a2b2a1a1b1a1a1 (G2a-P303>L497>L43>L42) sample in the Verteba Cave of Cucuteni-Trypillian culture which is dated 3619-2936 BCE, in modern southwestern Ukraine, with two other G2a-P303s.".

https://en.wikipedia.org/wiki/Bilche_Zolote#Verteba_and_Priest's_Grotto_Caves

https://www.yfull.com/tree/G-L43/

S04ancap
12-31-2019, 01:37 AM
There seem to be more, actually. At least Neolithic samples from the Vinca and Cucuteni-Trypillian cultures and a 800 BC Hallstatt culture sample: https://anthrogenica.com/showthread.php?9495-Origins-of-haplogroup-G-L497

The Vinca sample you linked to is actually the Baden Z1815, and the L42 from Ukraine ended up being removed feom the paper and given the dating wouldn’t even have been L42. The 800 BC Mitterkirchen sample may have been L497, but it was only 12 STR and no SNP’s so unconfirmed.

S04ancap
12-31-2019, 01:41 AM
In the study of ancient DNA "The Genomic History Of Southeastern Europe", there is a G2a2b2a1a1b1a1a1 (G2a-P303>L497>L43>L42) sample in the Verteba Cave of Cucuteni-Trypillian culture which is dated 3619-2936 BCE, in modern southwestern Ukraine, with two other G2a-P303s.".

https://en.wikipedia.org/wiki/Bilche_Zolote#Verteba_and_Priest's_Grotto_Caves

https://www.yfull.com/tree/G-L43/

This sample was removed from the subsequent paper and was probably contaminated and you can see the dates are way off as yfull shows L42 forming 1,700 BC and the sample is from 2,000 years earlier.

ArmandoR1b
12-31-2019, 12:50 PM
It looks like he may be G-L497 which would be an awesome find given the date of 4600 ybp and only the second BC L497 sample we have (the other being the Z1815 Baden sample from Hungary).

There is no coverage of the specific SNP G-L497 position 17423320 in the BAM file. It's phylogenetic equivalent of SNP S26003/Z3528 position 23971486 is derived but it is a G->A mutation which means it is most likely due to deamination especially since there are no derived SNPs between F1189/M3267/PF2883 and S26003/Z3528.

S04ancap
12-31-2019, 06:51 PM
There is no coverage of the specific SNP G-L497 position 17423320 in the BAM file. It's phylogenetic equivalent of SNP S26003/Z3528 position 23971486 is derived but it is a G->A mutation which means it is most likely due to deamination especially since there are no derived SNPs between F1189/M3267/PF2883 and S26003/Z3528.

So is all we know G-P287 like the paper says or is there something further downstream? I saw L497 online in the Ancient G-M201s with sequencing spreadsheet. https://docs.google.com/spreadsheets/d/1zs0UFYFHAwLTZu4mjm-nsylHxdkUzIWAT1VX4aWxMtk/htmlview

ArmandoR1b
01-01-2020, 04:01 PM
So is all we know G-P287 like the paper says or is there something further downstream? I saw L497 online in the Ancient G-M201s with sequencing spreadsheet. https://docs.google.com/spreadsheets/d/1zs0UFYFHAwLTZu4mjm-nsylHxdkUzIWAT1VX4aWxMtk/htmlview

Yes, all we know is he is derived (positive) for two SNPs that are phylogenetically equivalent to G-P287 which are F1239/M3483/PF2886 and F1189/M3267/PF2883 but one of those, F1239/M3483/PF2886, can be due to deamination. There is nothing further downstream that is derived that isn't possibly caused by deamination. The four derived SNPs downstream possibly caused by deamination contradict each other since they are in completely different subclades of G2. See below.



pos
marker_name
haplogroup
mutation
anc
der
reads
called_perc
called_base
state


8482393
F1239/M3483/PF2886
G2
C->T
C
T
1
100
T
D


8427005
F1189/M3267/PF2883
G2
A->G
A
G
1
100
G
D


23765526
Z45579
G2a2b1a1a1b
C->T
C
T
1
100
T
D


23971486
S26003/Z3528
G2a2b2a1a1b
G->A
G
A
1
100
A
D


15946166
FGC63520
G2a2b2a1a1c1a1a4
C->T
C
T
1
100
T
D


17612830
FGC65107
G2a2b2a1a1c4b
C->T
C
T
1
100
T
D