search this blog

Tuesday, September 25, 2018

AmtDB: an interactive ancient human mitogenome database


A very useful resource called AmtDB has just come online. For background info, check out the relevant paper by Ehler et al. here. Below is the paper abstract:

Ancient mitochondrial DNA is used for tracing human past demographic events due to its population-level variability. The number of published ancient mitochondrial genomes has increased in recent years, alongside with the development of high-throughput sequencing and capture enrichment methods. Here, we present AmtDB, the first database of ancient human mitochondrial genomes. Release version contains 1107 hand-curated ancient samples, freely accessible for download, together with the individual descriptors, including geographic location, radiocarbon dating, and archaeological culture affiliation. The database also features an interactive map for sample location visualization. AmtDB is a key platform for ancient population genetic studies and is available at https://amtdb.org.

To give an example of how this thing works, I'll search for a very specific mitochondrial (mtDNA) haplogroup, H6a1b, which was recorded, perhaps unexpectedly, in a sample from Hittite era Anatolia (individual MA2208 from Damgaard et al. 2018). I say perhaps unexpectedly, because it's a marker that is today, by and large, restricted to Northern Europe. Here are the results...


Interestingly, H6a1b only pops up in Copper and Bronze Age individuals from what are now Czechia, Great Britain, Poland and Russia, with not a single instance from the Near East. Moreover, the oldest sample on the list is from an Yamnaya culture burial in Samara, Russia. Thus, if the presence of this marker in the Hittite sample isn't due to contamination or poor quality sequencing, then it's likely that some Hittites belonged to mtDNA haplogroups that arrived in Anatolia from the steppes of what is now Russia.

See also...

Focus on Hittite Anatolia

25 comments:

Samuel Andrews said...

H6a1b & H6a1a are Steppe markers almost everywhere but maybe not everywhere.

Another thing to mention is that this database provides mtDNA raw data for lots of samples from the mega studies published in the last few years from which we only had mtDNA haplogroup labels.

This gives more complete information on mtDNA. The mtDNA haplogroup labels usually contain a few mistakes here and there anyways.

For example, this is boring information, but now we can see the Ukraine hunter gatherers had super inbred mtDNA and also that all Ukraine Mesolithic mtDNA types are found in later Neolithic hunter gatherers.

Ryan said...

"Interestingly, H6a1b only pops up in Copper and Bronze Age individuals from what are now Czechia, Great Britain, Poland and Russia, with not a single instance from the Near East. Moreover, the oldest sample on the list is from an Yamnaya culture burial in Samara, Russia."

So clearly Bell Beaker folk were WHG/EEF men with steppe women. :3

Lee Albee said...

Interesting database.
Thanks for sharing it.

I see what your saying with the H6a1.

But H6, H6a1 is not really well represented in the record. Trying to link it to the Eastern Europe as a source and Yamnya is probably a bit premature. Though it certainly could have been spread by them.


H6a1 is pretty common in Ashkenazi Jews-which definitely have near eastern roots--But also been linked to the Corded ware(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3806353/) so that might support your supposition.

However, H6a1 is dated to around 6764.5-10574.9 which suggests a neolithic expansion time frame. Origin could have been literally anywhere. Anatolia, Caucasus, Balkans......

Again the data is much too thin to make any sort of statement on origin

Lee


Ric Hern said...

Thank you very much.

Samuel Andrews said...

There's a lot of Mathieson 2018 data from western Europe I've never seen before. They don't have the new Andronovo/Sintashta data from Narasimhan 2018 though.

A thorough comparison could be done between Northern Bell Beaker and Andronovo anyways. That'd be cool mtDNA comparison because they represent the western and eastern ends of Bronze age IE expansions & both pops are a unique mixture of similar European Pastoral & farmer populations.

Samuel Andrews said...

@Lee Albee,

You're right. H6a1b in BA Anatolia isn't a smoking gun. Butt...

"However, H6a1 is dated to around 6764.5-10574.9 which suggests a neolithic expansion time frame. Origin could have been literally anywhere. Anatolia, Caucasus, Balkans......"

We need more than age estimates to know when haplogroups expanded. Age estimate+phylogeography=Expansion time & place.

Even if phylogeography confirms the expansion happened around 6ky, this expansion may have originally occurred in a small area then thousands of years later became widespread.

"Again the data is much too thin to make any sort of statement on origin"

The absence of H6a1 means as much as the presence. For example, H6a1 is absent in 100s & 100s of samples from Neolithic Europe but many examples exist in Steppe-admixed Bronze age Europe. Also, there are examples from Yamnaya, Afanasievo, Srubnaya, Sintashta. That basically confirms an association between H6a1 & 'Steppe.'

Davidski said...

@Lee Albee

I don't understand your point about the age of H6a1. As I see it, there's no difference here if H6a1 is 6,000 years old or 10,000 years old. Right now, with the available data, it really looks like H6a1b arrived both in Northern Europe and Anatolia during the Bronze Age with Yamnaya-related population movements.

EastPole said...

Very uesful tool for mtDNA research.
I checked H5a1

H5a is found in Neolithic Croatia and in Tripolye culture in Ukraine, then H5a1 pops up in CWC Estonia and in West European BB, next we find it in Karasuk culture in Russia:
https://i.postimg.cc/QMvmw0G0/screenshot_438.png

H5a1 most likely originated in CWC somewhere around Poland as it is now found predominantly among Slavic populations of Central and East Europe, including contemporary Poles:

http://eurogenes.blogspot.com/2017/12/descendants-of-ancient-european-fair.html?showComment=1512158992806#c3093765858090607319

It also found among Pamiri Tajiks and in India and Pakistan:

http://eurogenes.blogspot.com/2017/12/descendants-of-ancient-european-fair.html?showComment=1512158992806#c3093765858090607319

So I think we can try to tell the story of H5a1 as follows:

1. H5a came to Northern Europe with Anatolian Neolithic Farmers via Balkans.
2. Was present in Tripolye culture in Ukraine and from there it got into CWC.
3. Mutation H5a1 occurred somewhere in Poland around the time when CWC originated there and from Poland with CWC it migrated north, west and east.
4. West migration ended in BB culture.
5. North migrations lead it to CWC Estonia
6. East migration with CWC ended in Andronovo/Karasuk.
7. From Andronovo it most likely went via Pamir to India.

Would be nice to find H5a1 in Ancient Greece too. Maybe not necessarily Mycenaean Greece but some Hyperborean Girls were reported by Herodotus in Delos.

Lee Albee said...

@Davidsky

I do not dispute that the data is trending that way as of now. But there is insufficient data to make any definitive conclusions. It is equally possible that this haplogroup has an origin in Anatolia or the caucuses or ancient Iran and the sample has just not been found to date. Ancient DNA from the ancient near East it's still relatively thin on the ground.

The age depth of the haplogroup indicates that its origin and its time of expansion is from Neolithic. We know that Anatolian energy in populations began expanding around the same time. Therefore an argument can be made that it ended up in the on Yammnaya region during the neolithic. The same time in which traces of early farmer ancestry arrived into the steppe.

Again the theory that it came from the steppe is plausible. However,the autosomal data also suggests that there is very little sharing from the steppe into Anatolia during the early Bronze Age. Whereas the autosomal data does support Neolithic Farmers into the steppe populations.

Either directions can be supported by the data at this point. Or it's something completely different.

Davidski said...

@Lee Albee

Yamnaya doesn't have any ancestry from Neolithic farmers from south of the Caucasus. It has farmer ancestry from Middle Neolithic European farmers, and H6a1 is missing from these and the preceding early Neolithic European farmers.

So it's unlikely that H6a1 moved from Anatolia to the steppe, and that the H6a1b in the Hittite sample is from a lineage that stayed behind. More likely it's from the steppe.

Kristiina said...

@EastPole

H5 has been found in Anatolia Neolithic but there is also H5 from Mesolithic Eneolithic transition in the Balkans: Balcan Meso-Ene transition Lepenski Vir Lepe15 c. 6000 BC H5.

The origin of H in general is not clear.

Samuel Andrews said...

Paleolithic mtDNA from southwest Asia will reveal mHG H's early history.

Samuel Andrews said...

Even in Bronze age Britain, there's more U5 than H. About twice as much U5 and half as much H in people autosomally identical to modern northern Europeans.

IMo, this means natural selection has pushed down mtDNA U5 frequencies & erased a strong hunter gatherer signature in European mtDNA. I think there's enough samples now to be confident of this.

Chad Rohlfsen said...

LOL H5a1 here

Kristiina,

I believe Lepenski Vir was even more Anatolian-shifted than Iron Gates.

Lee Albee said...

@Davidski

Or it originated in GAC/LBK like populations. Then moved west, both into anatolia and into steppe.

Such that the minor steppe signal in one hittite is not steppe at all but shared population that fed into both??

Another possibility


Kristiina said...

@ Chad

Maybe.

The origin of H5 becomes clearer when we get 15 kya-10 kya H5 samples.

According to Soares, the age estimate of H5 is 6400-16800 and the point estimate is 11500. H6 is somewhat older: 10500-20300.

Both seem to have originated after the LGM.

Samuel Andrews said...

A lot of info is waiting to be unveiled from the rich ancient European mitogenome sample set we now have. I'm close to getting extremely thorough info on modern European mtDNA. After I do that, the ancient mtDNA will be easier to make sense out of. There appear to be note worthy differences between Bell Beaker & modern northwestern Europeans.

IMO, Andronovo mtDNA might show that Steppe populations were diverse similar to how Neolithic Anatolians were. Andronovo has a huge load of basal U2e1 and U4a1 that Yamnaya/Bell Beaker lack. Yamnaya has basal U5a1 that Andronovo/Beaker lack. This variation in U subclades is old. It isn't due to recent founder effects, it could be due to different EHG ancestors.

Andronovo's mtDNA is much more uniform & simple than Bell beaker. Andronovo looks like a population. Bell Beaker's mtDNA changes from site to site. It's hard to find trends.

Farmer mtDNA is weak in Andronovo but very significant in Bell Beaker. LBK/Danubian farmers are not the farmer ancestors of Bell Beaker. Not at all. IMO, the Danubian farmers might not even be the ancestors of Globular Amphora.

I don't just think this because typical Danubian lineage mHG N1a1a disappears in the Bronze age. The mHG H, K, T clades in bell beaker are also different from Danubian farmers.

Lee Albee said...

@Samuel Andrews

Try looking at the database for modern DNA here: https://empop.online/

you will need to ask for permission, and it has gaps but it is really interesting for visual assessment.

Really look to the rare mitogenomes as they will be most informative for migrations.

It is not seen in the ancient DNA record, but the pattern for X2b possibly suggest a paleolithic dissemination--possibly pre-WHG prior to 14K population replacement of El-miron cluster with Villabruna cluster

X2m looks like it may have been spread by Med route Neolithic farmers and X2n possibly by farmers going to the north.

All of it is very thin though datawise.

Family tree mtDNA maps seem to align with this though



For example the subhaplogroups for X2 look really interesting when combined with some ancient DNA

For example X2c. You have a split in X2c1 and X2c2, one is in southern europe/Iberia and the other seems to be Northeast Europe/scandinavia. Which suggests that one neolithic/mesolithic population in the NE had X2c with one group going the southern route via the med and one going the northern route into central Europe.

Open Genomes said...

@David and everyone, please don't forget that the Samara Enolithic sample I0444, R1b1a1a2a2 (Z2103*), Yamnaya Culture, Kutuluk, Samara, 3300-2700 BCE was H6a1b.

MA2208, who died in the attack on Kalehoyuk in 1750 BCE, Y G-M406, was in fact H6a1b2e, not just H6a1b.

H6a1b2e is found today only in Ireland and Denmark:
Genbank by Haplogroup: H6a1b

H6a1b2e is not the result of a Near Eastern or Caucasus migration northward to the Steppes. It's native to the Steppe.

This steppe origin of MA2208's mtDNA H6a1b2e is corroborated by the 7.4% "Afanasievo-like" ancestral component in this restricted nMonte3:


Restricted nMonte3 ancestry composition of sample: MA2208 Population: Anatolia_MLBA_low_res excluding Mycenaeans


With the Mycenaeans and other Iron Age samples included, we get 6.8% Mycenaean, 6.4% Baltic Bronze Age, and Shamanka Neolithic from Siberia:

Restricted nMonte3 ancestry composition of sample: MA2208 Population: Anatolia_MLBA_low_res incuding Mycenaeans

This is a very interesting combination. Northern and Eastern Steppe affinities, aside from perhaps Balkan ancestry, but only 10% Armenia Early Bronze Age. No direct CHG.

MA2208 seems to be of mixed Assyrian / Levantine ancestry on his paternal side, and some kind of Early Steppe Indo-European-Balkan ancestry on his maternal side. This "half" is our best look at an Early Indo-European Anatolian.

Samuel Andrews said...

@Lee Albee.

Yes. Thanks.

Lee Albee said...

@Open Genomes

The presence of H6a1b in the Samara Eneolithic sample does not mean it came from the steppe.
The earliest found location does not origin make.


"H6a1b2e is found today only in Ireland and Denmark" again the database your looking at is very small, and is likely very incomplete. Even if this was true, for today, it is still too thin to establish origin. Both areas have HG, neolithic and Bronze age ancestors that are likely of common origin. It only take one female to move the mtDNA haplogroup around.

Your nMonte look way overfitted, and have populations that most likely contain related ancestry. I am dubious about the reliability of the 7% ancestry from Afanasievo.

These nMonte do demonstrate a large amount of central european farmer ancestry and Anatolia and near eastern/caucus/Iran heritage. Which i do agree is likely. Hence that being a possible origin for the H6a1 mtDNA

Ric Hern said...

@ Open Genomes

Thanks for that info. Very interesting.

Samuel Andrews said...

@Lee Albee,

The association between H6a1 & Steppe in ancient DNA is pretty overwhelming. For modern Europe, we basically know for a fact their H6a1 is from the Steppe. Ultimately, all of Steppe's mHG H is of 'Near Eastern' (Paleolithic Caucasus hunter gatherer?) origin. Possibly H6a1 was born in the Near East making it possible this H6a1 in Anatolia is not of Steppe origin.

Davidski said...

@Samuel Andrews

Possibly H6a1 was born in the Near East making it possible this H6a1 in Anatolia is not of Steppe origin.

This doesn't make any sense, because the H6a1b2e in Bronze Age Anatolia is closely related to that in Northern Europe.

The H6a1b2e in Northern Europe does not come from early European farmers or from Anatolia. It comes from the steppe. The ancient and modern DNA are clear about that.

So either the H6a1b2e in Northern Europe and Bronze Age Anatolia both come from the steppe, or you have to argue that it arose in Anatolia via a parallel mutation process, which isn't possible.

Lee Albee said...

Do you have a paper that demonstrates this? I have not seen one , so would be willing to read more.