search this blog

Wednesday, September 3, 2014

First I1-M253 from prehistoric Europe


A preprint at bioRxiv reports on ancient DNA from the Balkans and Carpathian Basin. STA stands for Starčevo Culture, LBK for Linearbandkeramik Culture, and LBKT for LBK in Transdanubia.

The haplotype of the Mesolithic skeleton from the Croatian Island Korčula belongs to the mtDNA haplogroup U5b2a5 (Dataset S3). The sub-haplogroup U5b has been shown to be frequent in pre-Neolithic hunter-gatherer communities across Europe [28–30,32,33,45,46]. Contrary to the low mtDNA diversity reported from hunter-gatherers of Central/North Europe [28–30], we identify substantially higher variability in early farming communities of the Carpathian Basin including the haplogroups N1a, T1, T2, J, K, H, HV, V, W, X, U2, U3, U4, and U5a (Table 1). Previous studies have shown that haplogroups N1a, T2, J, K, HV, V, W and X are most characteristic for the Central European LBK and have described these haplogroups as the mitochondrial ʻNeolithic packageʼ that had reached Central Europe in the 6th millennium BC [36,37]. Interestingly, most of these haplogroups show comparable frequencies between the STA, LBKT and LBK, comprising the majority of mtDNA variation in each culture (STA=86.36%, LBKT=61.54%, LBK=79.63%). In contrast, hunter-gatherer haplogroups are rare in the STA and both LBK groups (Table 1).

...

Three STA individuals belong to the NRY haplogroup F* (M89) and two specimens can be assigned to the G2a2b (S126) haplogroup, and one each to G2a (P15) and I2a1 (P37.2) (Dataset S3, S5). The two investigated LBKT samples carry haplogroups G2a2b (S126) and I1 (M253). Furthermore, the incomplete SNP profiles of eight specimens potentially belong to the same haplogroups; STA: three G2a2b (S126), two G2a (P15), and one I (M170); LBKT: one G2a2b (S126) and one F* (M89) (Dataset S5).

So nothing really surprising there, as far as I can see. Y-haplogroup G2a has been found in plenty of other European Neolithic remains, so its status as the main Y-haplogroup of the men who introduced the Neolithic package from the Near East into Europe remains unchallenged. The I1 and I2 probably belonged to the descendants of indigenous European foragers who were incorporated into the ranks of the early farmers as they made their way north from Anatolia.

But the I1-M253 is interesting, because this is the first time that this haplogroup has been reported from prehistoric remains. I probably don't have to remind anyone that it's the most common Y-haplogroup in Scandinavia today, and yet it was missing from two sets Scandinavian forager remains analyzed recently by Lazaridis et al. and Skoglund et al. This perhaps suggests that it did not enter Scandinavia until the Neolithic, or even later.

Citation...

Anna Szécsényi-Nagy, Guido Brandt, Victoria Keerl, et al., Tracing the genetic origin of Europe's first farmers reveals insights into their social organization, bioRxiv, first posted online September 3, 2014, doi: http://dx.doi.org/10.1101/008664.

24 comments:

Richard Rocca said...

Most shocking is the continued lack of Y-haplogroup J in any farming aDNA study.

Davidski said...

I suspect this guy's a potential J man.

http://penn.museum/press-releases/1093-ur-skeleton-rediscovered.html

Chad Rohlfsen said...

Not really one to toot my own horn, but I1 is spread by a neolithic culture, as I said. Nothing to do with any mesolithic survival or Germanic migration.

Shaikorth said...

These results show that I1 and I2 existed in farmer cultures to some extent, but also that the main lineages spread by them are clearly G2a and F, which are no longer around in Europe in relevant numbers. I1 clearly existed independently from the G2 and F farmers elsewhere as we know I2 did, otherwise it would be accompanied by other farmer lineages in its range.

The more southern and earlier Starčevo remains don't have any I1 but lots of G2 and F, which is telling.

Lathdrinor said...

The paper does state that E1b was found at a site in Spain, albeit it's an aberration across the mid-Neolithic landscape in Europe. The rest of the missing Y haplogroups, J and R specifically, look firmly tied to the late Neolithic and later - an event horizon for populations across Eurasia.

Chad Rohlfsen said...

I'll bet j's are the island hoppers.

Fanty said...

Another interesting thing is, that its an I1, that predates the suposed MRCA of all modern I1 (wich was calculated to only 5000 years) by 2000 years.

What does it mean?
Does it push back the MRCA to that date? Or is this a I1 lineage that did not make it to the present or does this special lineage pass through that single person bottleneck event 5000 years ago?

barakobama said...

"Not really one to toot my own horn, but I1 is spread by a neolithic culture, as I said. Nothing to do with any mesolithic survival or Germanic migration."

Most people including myself suspected I1 was a originally a central European hunter gatherer lineage, that was absorbed by near eastern farmers and then finally by Indo Europeans. It's obvious that LBK got I1 from hunter gatherers just like they got I2a1-P37 from hunter gatherers. Farmers had been in that area of Europe for less than a thousand years at that time, so I highly doubt unlike other I lineages that it was of farmer origin.

Now we know I1 existed in Neolithic and Mesolithic central Europe(and probably other regions), and that it was probably much less popular than it is today in central Europe and Scandinavia. I1 was largely spread by Germanic people who orignated in a small area in southern Scandinavia(It's existence in eastern Europe, central Europe, and Britain are prove).

I1 though is pretty diverse in Europe today and there are no obvious very young founder effects like with R1a and R1b, so maybe it's popularity is somewhat of a Mesolithic-Neolithic relic but I really doubt it.

Not many people payed attention to Genetinker's Y DNA results for PWC hunter gatherers and Gok4(he had I2, obviously probably I2a1-P37). Ajv52 very likely had I2a2a1. So, there's already prove now that the main hg I lineages of modern northwest Europeans descend originally from Mesolithic ones.

Daniel Szelkey said...

@Richard Rocca why would J be in ancient Europe? Its stuck in areas where phoencians lived. I am surprised by lack of E1b1b but not J.
@Chad Rohlfsen The I y-dna in the linearback means that the linearback had male meditated mesolithic european gene flow as around 10-20% seemed to have I but less than 2% had U5/U4/U2 . This means that some Linearback farmer girls mated with European hunter gatherers and were still accepted into linearback society and that linearback men were not allowed or able to take hunter women till latter. This means that the component EEF is complete baloney as it is probably around 10% WHG admixture, and this proveS my point that the Sardinians have at least 10% WHG blood.

Daniel Szelkey said...

All of the F* is G2, this is low resolution and years ago Nasizde confused F with G2 in the Caucasus. The same is happening with ADNA. I am absolutely certain of this.

Grey said...

@Fanty

"Another interesting thing is, that its an I1, that predates the suposed MRCA of all modern I1 (wich was calculated to only 5000 years) by 2000 years. What does it mean?"

I think it will eventually turn out to mean that only very few hunter-gatherers made the forager to farmer transition (probably in herder form) thus creating extreme bottle necks in their y dna and then one or more of these survivor populations tagged along on later expansions like the mixed I & R Germanic.

(I think it might tie into the theory that the Vanir & Aesir in Norse theology are the gods of two separate populations that joined together.)

(Alternatively, given the remote and often mountainous distribution of I maybe a caste of neolithic miners.)

Chad Rohlfsen said...

@BO,
You can't know for sure that I1 was present in Mesolithic men, at all. For all we know it came from a Neolithic man that had I*. You can't know for certain, until we find a Mesolithic person that is I1.

@Daniel
Sardinians tested by Lazaridis, ranged from 5.8-18.2% WHG.

barakobama said...

"You can't know for sure that I1 was present in Mesolithic men, at all. For all we know it came from a Neolithic man that had I*. You can't know for certain, until we find a Mesolithic person that is I1. "

People said the same about I2a1-P37 when the only ancient samples were from Neolithic France. I remember Maju and Davidski argued that hg I itself came to Europe from the near east in the Neolithic using the Neolithic French samples as evidence, a few weeks before Laz 2013.

Ancient DNA has given many surprising and groundbreaking results(ANE ancestry?) no one could have perfectly predicted and that contradict the best theories based on modern genetics, but I'm still going to go along with the mainstream idea based on modern Y DNA that I1 and or pre-I1 existed in Mesolithic central Europeans.

It's obvious I1 was originally a hunter lineage. It has many mutations passed I*, and we don't know if this Neolithic I1 had all of the mutations modern I1's do. If modern I1 didn't originate in Euro hunters, pre-I1 did.

I agree with you that some I1 in Europe today truly is a Neolithic and Mesolithic relic. I made that argument last year, when everyone assumed it was a young Germanic-specific lineage. It was obvious to many people the Germancentric theory was to simplistic because central Europeans have a differnt ratio of I1 clades than Scandinavians, Finnish-specific clades, and some clades like I1b have only been found in central Europe. Another problem with the argument that I1 is a young founder effect, is how diverse it is, and now I1 ~7000YBP in central Europe's first farmers.

This study is old school because it only got weak mtDNA-Y DNA results. So, their testes of relatedness of the ancient samples with modern populations is very limited and prone to inaccuracies and false relatives.

Going off Y DNA and mtDNA haplogroup percentages, as evidence for a lineage's place of origin, predicting amounts of ancient ancestry, and testing relatedness between populations, has been proven to be unreliable many times. Yet this study like many old studies trusted their weak mtDNA and Y DNA results to help learn how these farmers relate to modern populations.

I think haplogroup percentages of very old lineages should be considered, but the goal obviously is to to break down the mtDNA or Y DNA sample as much as possible. Working with haplotypes is the key. Knowing a LBK sample had H is pretty much useless information. We need to know what SNPs were tested, what H clade's mutations it lacks, what it's the possible clades it belongs to are, and what it's extra mutations are.

barakobama said...

My great great grandfather, Olaf was born at the coast of the North sea in Norway belonged to I1, so i kind of consider it one of my haplogroups.

barakobama said...

" The I y-dna in the linearback means that the linearback had male meditated mesolithic european gene flow as around 10-20% seemed to have I but less than 2% had U5/U4/U2 . This means that some Linearback farmer girls mated with European hunter gatherers and were still accepted into linearback society and that linearback men were not allowed or able to take hunter women till latter. "

It seems some people assume the hunters were brutally masculine and the farmers were soft and only good with brains(in reality they were tough people), and admixture was usually hunter man+farmer woman. I think that's a mostly inaccurate stero type. There was probably no gender trend, just random admixture.

I don't think we have enough samples to conclude how they mixed with each other. Maybe your on to something that I refuse to take seriously.

Dr Rob said...

Grey said
"I think it will eventually turn out to mean that only very few hunter-gatherers made the forager to farmer transition (probably in herder form) thus creating extreme bottle necks in their y dna and then one or more of these survivor populations tagged along on later expansions like the mixed I & R Germanic."

Exactly, good point (although your bracketed points are unlikely;) )

This goes to show the folly in trying to determine the "ages" of haplogroups based on STR accumulations etc

Maybe Zhivotovsky et al did have a point afterall, wrt the "effective evolutionary rate "

Maju said...

@Shaikorth: G2a is not that rare: it includes around 5% of all Iberian lineages (with peaks in Ibiza (13%), Portugal (9-12%) and Southern Iberian Plateau (10%)), some 11% of Italian ones and 14% in Sardinia. Other places with important (>5%) G/G2a frequencies in non-Caucasian Europe are: Provence (8%), Tyrol (7%), some Croatian islands (6-10%), Munich (5%), Macedonian Slavs (5%) and Romanians (5%). It may also be the case in Crete (11%) and the Orlovska Oblast in Belarus (10%) (cf: http://en.wikipedia.org/wiki/Haplogroup_G_%28Y-DNA%29_by_country - notice that G in non-Caucasian non-Jewish Europeans is almost invariably G2a).

This implies an important level of continuity of Neolithic patrilineages in Southern Europe, very especially in Italy and Sardinia (unsurprising, really), although clearly diluted in any case.

I am more and more intrigued about Y-DNA J2b, which is also important in about the same areas of Europe where G2a is (~8% in Iberia, 9% in Sardinia, 20% in Italy, also high in Crete and Thessaly) but which has not been detected so far among early farmers. I even wonder if it may be some sort of Italian-specific founder effect which was later spread by Romans to Iberia (or who knows!)

Maju said...

@Fanty:

Another interesting thing is, that its an I1, that predates the suposed MRCA of all modern I1 (wich was calculated to only 5000 years) by 2000 years.

What does it mean?


It means that your usual molecular-clock-o-logy is a massive fail, pseudoscientific nonsense with a meaningless academic (scholastic) varnish. It means that the usual STR-based Y-DNA MC age guesstimates are all way too recent by a lot.

Said that, it's possible that using full chromosome sequencing and proper calibration points, realistic estimates may be achieved. Some efforts have been done in this direction but they are still shy and rare. But just because someone with a degree throws a hunch with some loose mathematical argumentation (and others follow suit out of scholastic inertia), that doesn't mean that it's true. Science is not built on authority but on critical thought, which is almost the opposite.

jackson_montgomery_devoni said...

Maju,

I think that within Europe today J2b may reach it's peak among Gheg Albanians but I may be wrong. Either way when it comes to Europe today it is found most frequently in the Balkans I believe.

Maju said...

Your're surely right, Jackson. I quickly browsed Wikipedia but I obviously missed the Albanian 19%. Also I recall from old sources that J2 is common in the Eastern Balcans (maybe 20%?)

On second thought, the most parsimonious explanation based on modern frequencies would be to dub it Neolithic. However the lack of it in ancient Neolithic samples is growingly unsupportive of this theory.

Speculatively speaking I could imagine that:

1. It arrived to the Balcans with the secondary "black pottery" wave (rel. to Tell Halaf and Can Hassan), best known for the Vinca culture, which I usually dub in the wider region as Dimini-Vinca (some phases of Karanovo also belong to it at least partly), which persisted in Greece until the arrival of the Indoeuropeans (Greeks proper in the ethno-linguistic sense).

2. Italy was influenced from the Aegean in the Chalcolithic and possibly even colonized in the Bronze Age (Etruscans notably). If this impact was very strong, current J2 levels can be explained out of it.

3. Either in some older (but mysterious) period from the Aegean or otherwise Eastern Mediterranean or in historical times from Italy, the lineage arrived to Iberia.

sblog said...

@Maju

The discovery of this I1 ancient DNA sample is based on only one marker M253. The definition of the I1 haplogroup on ISOGG is based on a list of many SNPs: CTS5783/S63, CTS11042/S66, L64, L75, L80, L81, L118, L121/S62, L123, L124/S64, L125/S65, L157.1, L186, L187, L840, L845/S110/YSC0000278, M253, M307.2/P203.2, M450/S109, P30, P40, S107, S108, S111/Y1868

If M253 SNP is one of the first markers of this list, it can be much older than the common ancestor of all the living I1 people.

Daniel Szelkey said...

It can be older, if the old dating is estimating ages too young.

Bernard said...

@Daniel Szelkey
Not necessary

Joe Flood said...

I noticed that I1 sample, it is very interesting. After being apparently stagnant for 23 millennia, I1 began its expansion at virtually the same time as R1a and Atlantic R1b - early Bronze 3000-2700 BC. It's quite possible it accompanied or followed R1b around from the Mediterranean, it only takes one man. The epicentre of the Bronze expansion appears to be in Denmark.