Revisiting Stora Förvar 11 from Gotland

[I1a3_Young]

The island of Gotland off the east coast of Sweden has a couple smaller islands on the western side, one of which is Stora Karlsö. The Stora Förvar cave is on that small western island. The Stora Förvar 11 sample was obtained in 1887.

Gotland.pngStora_Förvar_Stora_Karlsö.jpg

The age is 5550-5300 BC and the context of the culture is Late Mesolithic.

When originally published, no definitive Y-DNA haplogroup was given. Stora Förvar 11 was noted as positive for a CT, which is ancestral to F and then I. I believe that this sample deserves more attention in our community because while the Y-DNA haplogroup was not published in a paper, hobbyists like us have determined a very significant fact:

Stora Förvar 11 was I1

This was determined by Genetiker but also corroborated by reputable posters on another forum. I took the data and confirmed it independently. Stora Förvar 11 definitely has I1 SNP calls. He is positive for:

Z2805/CTS6629

Z2802/CTS6221

Z2792/CTS5887

YSC0000301/Z2882

Z2726

Z2749/CTS1748

Z2731

https://genetiker.wordpress.com/y-snp-calls-for-stora-forvar-11/

This is very significant due to the lack of ancient I1 samples. Previously I had only read of BAB5 from the LBK culture in Hungary. What's even more interesting is that this sample is estimated to be contemporary to BAB5

So we have the earliest I1 confirmed at the same time from Hungary to Gotland (Sweden). That is a rather large spread. Was there a lot of I1 in this area or were these outliers?

The autosomal data is equally exciting. http://linearpopulationmodel.blogspo...or-motala.html

Stora Förvar 11 is 72% Euro HG, 8% EEF, 17% N. Caucus while BAB5 is 90% EEF and about 10% Euro HG. The above link shows a comparison with the Motala hunter/gatherers who were I2a1 and pure Euro HG. So we not only have a large geographical spread of ancient contemporary I1 but also a large autosomal gap, even though both samples do contain EEF components. In fact, all I1 samples that I've seen contain an EEF component, while we have many I2 samples with zero EEF. For both SF11 and BAB5, the small 8-10% component indicates either an older mixture of the EEF and Euro HG population or a direct mixing 4-6 generations back.

The mtDNA were U5a1 for SF11 and H for BAB5.

Does the 17% N. Caucus also raise eyebrows? I thought that component didn't enter Europe until IE expansion, but I don't know enough to make this determination (thus this thread).

I'm going to include this along with additional info sometime in the I1 section in the future. I put this in neolithic instead of paleolithic and mesolithic because it's just before the farming spread to the area and the sample contained a portion of EEF DNA.

[bicicleur]

SF11 age and autosomal DNA are similar to the Motola samples
SF11 is pré-I1 but not ancestral to I1 : it is poisitive for some I1 SNP, but also negative for other
The age 5550-5300 BC compared to TMRCA of I1 means that it should be positive for at least 85 % of the SNP in order to be ancestral to I1.
https://www.yfull.com/tree/I1/

the fact that both BAB5 and SF11 were pré-I1 suggests that pré-I1 was spread as a minority to I2 all over Europe, from the Baltics to the Balkans

[I1a3_Young]

SF11 age and autosomal DNA are similar to the Motola samples
SF11 is pré-I1 but not ancestral to I1 : it is poisitive for some I1 SNP, but also negative for other
The age 5550-5300 BC compared to TMRCA of I1 means that it should be positive for at least 85 % of the SNP in order to be ancestral to I1.
https://www.yfull.com/tree/I1/

the fact that both BAB5 and SF11 were pré-I1 suggests that pré-I1 was spread as a minority to I2 all over Europe, from the Baltics to the Balkans

The definition of I1 is based on modern humans as we know them. Rather than taking the earliest split of I1 from I2 and designating all I1 as such, the community seems to call it "pre-I1" because I1 is more modern.

There were thousands of years worth of I1 branches but almost all of them went extinct, until the DF29 population explosion. We define I1 by what modern people have rather than all ancient I1 branches.

I think it is more correct to label all ancient I1 as I1 rather than pre-I1 because they are certainly part of the I1 umbrella, but not part of the unique and popular surviving branch.

I1-Z2864/L509
I1-Z2805/CTS6629
I1-Z2802/CTS6221
I1-Z2825/CTS9258
I1-Z2792/CTS5887
I1-L121/S62
I1-YSC0000301/Z2882
I1-Z2762/CTS3268
I1-Z2726
I1-Z2749/CTS1748
I1-Z2804/CTS6547
I1-Z2690
I1-CTS9845
I1-Z2731
I1-FGC2425/Z2714
I1-FGC7871/Y1906
I1-FGC2468/Y1831

That's 7 positive and 10 negative for the pre-DF29 calls. I'm not sure on the ages of the last three.

The only call I'm aware of for BAB5 is I-M170 and I-M253 for groups I and I1.

It has to be likely for large variation of pre-DF29 I1 due to the huge amount of time between I1 genesis and DF29 genesis.

The only real ancestral I1 we can dig up will be Mr. DF29's father line. The farther away a sample is to him, the more I1 calls will be negative.

[I1a3_Young]

https://genetiker.wordpress.com/2014...var-11-genome/

That is genetiker's analysis but after checking his blog I would be wary of certain biases on any of his results.

Is there a good source for Motala vs SF11 other than the one I posted which showed them to be very similar but with small key differences?

[Alpenjager]

Is there a GEDmatch number for BAB5?

[Fire Haired14]

I think it is more correct to label all ancient I1 as I1 rather than pre-I1 because they are certainly part of the I1 umbrella, but not part of the unique and popular surviving branch.

I think we should call Stora Forvar pre-I1 because I1 is the name of the popular surviving branch and Stora Forvar doesn't belong to that branch. Imo, either I1 should be renamed to I1a and Stor Forvar named I1 or I1 should remain I1 and Stora Forvar remain pre-I1.

Lots of pre-haplogroups will be found in ancient DNA, so we'll run into this problem many more times in the future.

We've already run into this problem with Paleolithic Y DNA/mtDNA. One example is the Paleolithic mtDNA pre-U5 from Europe. Geneticists responded by calling the pre-U5 in UP Europe U5, and calling modern U5 U5a'b. That's the equivalent of renaming modern I1 I1a and naming Stora Forvar I1.

[I1a3_Young]

I suppose that is a fair point. The structure of pre-modern I1 is a mystery so we can't discuss any branches. However, the significance of pre-I1 should not be minimized by some in our community. This SF sample being Pre-I1 is huge. I suppose if we all understand that Pre-I1 also means Post-I then we're good.

Think of how hard R1b discussions would be if you had to refer to all pre M269 as Pre-R1b.

[I1a3_Young]

GEDmatch for SF11: Z466794

MDLP K11: 68% WHG, 15% EHG, 5% EEF, rest African/other
Eurogenes K13 Oracle: Baltic
Dodecad V3 Oracle: Finnish/Swedish

Mlukaz posted a comparison of Motala and SF11 in Eurogenes K36 showing they are mostly similar. However, in that comparison SF11 was significantly less "Fennoscandian" and more "eastern europe." That fact, combined with the numbers above, brings up the following question for me:

Was SF11 a mix of native Nordic HG with pre-Corded Ware baltic area people? It seems very plausible that the Baltic coast people would navigate and spread around. The Motala HGs were in central inland Sweden and would have been more slow to be mixed. If this is the case and SF11 has "old Baltic" autosomal DNA, then is the "pre-I1" from the Swedish HG side or the Baltic side? The mtDNA appears like it could be either.

[Lukas]

GEDmatch for SF11: Z466794

MDLP K11: 68% WHG, 15% EHG, 5% EEF, rest African/other
Eurogenes K13 Oracle: Baltic
Dodecad V3 Oracle: Finnish/Swedish

Mlukaz posted a comparison of Motala and SF11 in Eurogenes K36 showing they are mostly similar. However, in that comparison SF11 was significantly less "Fennoscandian" and more "eastern europe." That fact, combined with the numbers above, brings up the following question for me:

Was SF11 a mix of native Nordic HG with pre-Corded Ware baltic area people? It seems very plausible that the Baltic coast people would navigate and spread around. The Motala HGs were in central inland Sweden and would have been more slow to be mixed. If this is the case and SF11 has "old Baltic" autosomal DNA, then is the "pre-I1" from the Swedish HG side or the Baltic side? The mtDNA appears like it could be either.

I'm converting to Gedmatch format Latvian and Estonian mesolithic / neolithic genomes from papers ad 2017. Shortly we will compare them with Motala nd Stora Forvar...

[LeBrok]

I'm converting to Gedmatch format Latvian and Estonian mesolithic / neolithic genomes from papers ad 2017. Shortly we will compare them with Motala nd Stora Forvar...

Right on, mlukas! IIRC the plotted closer to WHG on PCA chart than Ukrainian samples.

[Alpenjager]

I uploaded them to GEDmatch some month ago:

Z229669 HG3 Latvia, M001884 HG1 Latvia.

I don't know if there is enough autosomal DNA from BAB5 to be uploaded to Gedmatch.

[LeBrok]

I uploaded them to GEDmatch some month ago:

Z229669 HG3 Latvia, M001884 HG1 Latvia.

I don't know if there is enough autosomal DNA from BAB5 to be uploaded to Gedmatch.

Is there a list of a thread on gedmatch with ancient samples so people don't duplicate work and can find samples they are looking for?

[LeBrok]

Latvia HG3 was a very decent sample, here it is when compared to other Mesolithic guys in HarappaWorld run.

M325047	KO1, I-L68		Z229669	HG3		T457572	Vsilyevka 3		F999917	I-L416		M218547	I0124		M643041	I0061
Hungarian, Tiszaszőlős-Domaháza	7.7 kya		Latvia	Mesolithic		Ukraine, Vasilyewka	Mesolithic		Motala 12 Östergötland, Sweden	7 kya		Samara HG	7.6 kya		Karelia, OleniyOstrov N Russia	7.25 kya
Run time	9.43		Run time	6.6		Run time 3.46	3.46		Run time	8.67		Run time	5.57		Run time	9.88
S-Indian	-		S-Indian	-		S-Indian	-		S-Indian	-		S-Indian	-		S-Indian	-
Baloch	-		Baloch	-		Baloch	1.92		Baloch	-		Baloch	14.33		Baloch	9.46
Caucasian	-		Caucasian	-		Caucasian	-		Caucasian	-		Caucasian	-		Caucasian	-
NE-Euro	80.37		NE-Euro	89.36		NE-Euro	84.32		NE-Euro	90.24		NE-Euro	75.62		NE-Euro	72.66
SE-Asian	-		SE-Asian	-		SE-Asian	-		SE-Asian	-		SE-Asian	-		SE-Asian	-
Siberian	-		Siberian	0.33		Siberian	2.03		Siberian	0.07		Siberian	-		Siberian	-
NE-Asian	-		NE-Asian	-		NE-Asian	-		NE-Asian	-		NE-Asian	-		NE-Asian	-
Papuan	0.53		Papuan	-		Papuan	-		Papuan	0.57		Papuan	-		Papuan	-
American	-		American	0.13		American	0.83		American	1.58		American	9.62		American	12.6
Beringian	-		Beringian	-		Beringian	-		Beringian	0.68		Beringian	0.15		Beringian	5.16
Mediterranean	18.59		Mediterranean	9.84		Mediterranean	10.51		Mediterranean	6.83		Mediterranean	-		Mediterranean	-
SW-Asian	-		SW-Asian	-		SW-Asian	-		SW-Asian	-		SW-Asian	-		SW-Asian	-
San	-		San	-		San	-		San	-		San	-		San	-
E-African	-		E-African	-		E-African	-		E-African	-		E-African	-		E-African	-
Pygmy	-		Pygmy	-		Pygmy	-		Pygmy	-		Pygmy	-		Pygmy	0.07
W-African	0.5		W-African	0.34		W-African	0.39		W-African	-		W-African	0.2		W-African	-

I wonder if Baloch component showed up late in a game. Late Mesolithic. It didn't transmit to Scandinavia and Latvia with EHGs. I guess, the components that made EHG first were mainly American and Beringian admixtures, and maybe a bit of Siberian, unless Siberian is a noise here. Baloch came later from South-East.

[Lukas]

I uploaded them to GEDmatch some month ago:

Z229669 HG3 Latvia, M001884 HG1 Latvia.

I don't know if there is enough autosomal DNA from BAB5 to be uploaded to Gedmatch.

But only those two Latvian you uploaded? I want to convert rest but must be sure not duplicate...

BTW what methods of conversion did you use?

[Lukas]

BTW on anthrogenica is thread for this http://www.anthrogenica.com/showthre...es-on-GEDmatch

[Lukas]

moved.....

[Lukas]

I uploaded

T332765	Latvia_MN1 (Zvej26)
T400485	Latvia_LN1 (zvej28)

In K36 Eurogenes
1,88 westafrican due to postmortem damage probably.

	T332765	T400485
	Latvia_MN1 (Zvej26)	Latvia_LN1 (zvej28)
Amerindian	0	0
Arabian	0	0
Armenian	0	0
Basque	3,91	0
Central_African	0	0
Central_Euro	0	7,67
East_African	0	0
East_Asian	0	0
East_Balkan	0	0
East_Central_Asian	0	0
East_Central_Euro	11,91	7,58
East_Med	0	0
Eastern_Euro	24,44	16,05
Fennoscandian	12,57	18,59
French	0	0
Iberian	0	0
Indo-Chinese	0	0
Italian	0	0
Malayan	0	0
Near_Eastern	0	0
North_African	0	0
North_Atlantic	20,44	13,97
North_Caucasian	0	4,1
North_Sea	26,72	30,17
Northeast_African	0	0
Oceanian	0	0
Omotic	0	0
Pygmy	0	0
Siberian	0	0
South_Asian	0	0
South_Central_Asian	0	0
South_Chinese	0	0
Volga-Ural	0	0
West_African	0	1,88
West_Caucasian	0	0
West_Med	0	0

[LeBrok]

I checked in Harappa run. MN1 looks very like WHG, almost identical to Hungarian sample KO1. LN1 is inconclusive do to bad sample quality, it could be either EHG or Yamnaya or something inbetween. Definitely though, most of its genome is made of EHG.

Is MN2 a decent quality to have a look?

[LeBrok]

I'm going to make a thread about ancient samples of Gedmatch.

[Lukas]

I checked in Harappa run. MN1 looks very like WHG, almost identical to Hungarian sample KO1. LN1 is inconclusive do to bad sample quality, it could be either EHG or Yamnaya or something inbetween. Definitely though, most of its genome is made of EHG.

Is MN2 a decent quality to have a look?

It's biggest of those two other MN. I will convert it by night:)

[Lukas]

I uploaded them to GEDmatch some month ago:

Z229669 HG3 Latvia, M001884 HG1 Latvia.

I don't know if there is enough autosomal DNA from BAB5 to be uploaded to Gedmatch.

Could you convert also this Estonian genomes from:
Extensive farming in Estonia started through a sex-biased migration from the Steppe (http://biorxiv.org/content/early/2017/03/02/112714)

Here, are the files: http://evolbio.ut.ee/CCC_CWC/

I tried to convert but I've got error about bad headers. Maybe they are really corrupted but I hope different tool will be enough.

[Alpenjager]

I have uploaded Z615442 HG1 and Z455885 N1 from Ukraine too together with other ancient genomes. Which genomes do you want to upload?

I converted BAM to readeable files, then I used CSVs to mount the TXT for upload them to GEDmatch

[Alpenjager]

Do you know IDs for each of these BAM files?

[Lukas]

Do you know IDs for each of these BAM files?

Somebody converted them and give me files. Those marked low are low-coveraged.

	Gedmatch
Kudrukula2_MA974_LOWx	Z015901
Kudrukula3_MA975_LOWx	Z046520
Kunila2_MA973	Z973823
Sope_MA826	Z033947
Ardu2_MA969	Z491053
Ardu1_MA968_LOWx	Z784415
Kunila1_MA971	Z607891