search this blog

Monday, June 19, 2017

Polish aDNA PCA

Below is a Principal Component Analysis (PCA) that I put together for an upcoming presentation on Polish ancient DNA (aDNA). The five RISE samples are from Allentoft et al. 2015, including RISE569, the early Slavic genome from the Czech Republic, which was initially wrongly labeled as a Czech Bell Beaker (see here). PL_N17 is an Early Bronze Age (EBA) sample from Gustorzyn, Northern Poland (see here).

I also organized f3 outgroup statistics of the form f3(European_pop,Test,Yoruba) for each of these samples to compare their genetic affinities to present-day European populations. Although f3 outgroup statistics aren't as sensitive as haplotype tests, I think these results look interesting and useful, with both PL_N17 and RISE569 seemingly showing strong links to modern-day West Slavs. The full output is available in a zip file here.

Poland_EBA PL_N17
Lithuanian 0.175778
Ukrainian_West 0.174866
Sorb 0.174334
Estonian 0.174313
Icelandic 0.17397
Irish 0.173863
Polish_West 0.173743
Polish_East 0.173549
Czech 0.173545
Norwegian 0.173533

Early Slav RISE569
Sorb 0.169171
Lithuanian 0.168945
Estonian 0.168819
Polish_West 0.168267
Polish_East 0.168143
Irish 0.168092
Czech 0.167941
Norwegian 0.167787
Icelandic 0.167696
Finnish 0.167685

See also...

Testing for genetic continuity in Poland from the Bronze Age to the present

Saturday, May 20, 2017

Shared maternal ancestry between Slavs and Germanics probably dates to the Metal Ages

Over at the Russian Journal of Genetics behind a paywall at this LINK. Emphasis is mine:

Abstract: The structure and diversity of mitochondrial DNA (mtDNA) macrohaplogroup U lineages in Russians from Eastern Europe are studied on the basis of analysis of variation of nucleotide sequences of complete mitochondrial genomes. In total, 132 mitochondrial genomes belonging to haplogroups U1, U2e, U3, U4, U5, U7, U8a, and K are characterized. Results of phylogeographic analysis show that the mitochondrial gene pool of Russians contains mtDNA haplotypes belonging to subhaplogroups that are characteristic only of Russians and other Eastern Slavs (13.7%), Slavs in general (11.4%), Slavs and Germans (17.4%), and Slavs, Germans, and Baltic Finns (9.8%). Results of molecular dating show that ages of mtDNA subhaplogroups to which Russian mtDNA haplotypes belong vary in a wide range, from 600 to 17000 years. However, molecular dating results for Slavic and Slavic-Germanic mtDNA subhaplogroups demonstrate that their formation mainly occurred in the Bronze and Iron Ages (1000–5000 years ago). Only some instances (for subhaplogroups U5b1a1 and U5b1e1a) are characterized by a good agreement between molecular dating results and the chronology of Slavic ethnic history based on historical and archaeological data.

Malyarchuk, B.A., Derenko, M.V. & Litvinov, The macrohaplogroup U structure in Russians, A.N. Russ J Genet (2017) 53: 498. doi:10.1134/S1022795417020053

Tuesday, May 16, 2017

Globular Amphora people starkly different from Yamnaya people

The figure below is from the recent Mathieson et al. 2017 preprint; slightly edited to highlight the results of nine Globular Amphora Culture (GAC) samples from two burial sites in what are now Poland and Ukraine.

Despite living in East Central Europe at about the same time as the nearby Yamnaya people of the Pontic-Caspian Steppe, these GAC individuals show practically zero Yamnaya-related or steppe ancestry (note the almost total absence of the orange "Yamnaya" component in the Globular_Amphora results in the ADMIXTURE bar graph). Instead, they're very similar to Chalcolithic and Middle Neolithic Central and Western Europeans, with whom they overlap in the Principal Component Analysis (PCA).

During the tail end of the GAC period, East Central Europe was suddenly dominated by a new archaeological complex called the Corded Ware Culture (CWC). Although most CWC individuals sampled to date show minor GAC-related ancestry, they're overwhelmingly Yamnaya-like, which suggests that by and large the CWC population has its origins on the Pontic-Caspian Steppe. In fact, some of the earliest CWC examples from the Baltic States, such as Latvia_LN in the ADMIXTURE bar graph, are basically identical to Yamnaya people.

It was suggested not long ago that the presence of Yamnaya-related ancestry in modern-day Europeans could be mostly explained by the so called Isolation-by-Distance phenomenon (see here). But as I said at the time, this was a major faux pas, and thanks to these GAC samples I now have direct evidence from ancient DNA to back me up. So forget the idea of anything resembling a gentle cline in Yamnaya-like ancestry east to west across Europe before proto-CWC and Yamnaya exploded from the steppes.

By the way, in that critique I said that it's not possible to recapitulate ancient populations with ADMIXTURE components. I stand by that statement, although as we can see in Mathieson et al. 2017, it is possible to get close at times with enough of the right ancient samples; close enough to make some general observations anyway.

Interestingly, on the PCA plot, the European Bronze Age cluster is more or less half way between GAC and Latvia_LN. This is also where modern-day Poles and Ukrainians cluster on such plots when they're not significantly skewed by projection bias or shrinkage. Thus, I do wonder if the Slavs of East Central Europe are essentially a 50/50 mixture of early CWC and late GAC? I'll try and test this when the Mathieson et al. 2017 dataset goes online.


Mathieson et al., The Genomic History Of Southeastern Europe, bioRxiv, Posted May 9, 2017, doi:

See also...

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

Tuesday, March 28, 2017

Hints of deep genetic substructure in Iron Age Poland

A paper at Infection, Genetics and Evolution looks at the susceptibility to infectious diseases in two late Iron Age groups from Central Poland. I can't wait to see genome-wide and Y-chromosome data from these and other ancient Polish populations. Judging by the outcomes presented in this paper, and also rumors that I've heard from Polish labs, we're in for some major surprises. Emphasis is mine:

Abstract: For thousands of years human beings have resisted life-threatening pathogens. This ongoing battle is considered to be the major force shaping our gene pool as every micro-evolutionary process provokes specific shifts in the genome, both that of the host and the pathogen. Past populations were more susceptible to changes in allele frequencies not only due to selection pressure, but also as a result of genetic drift, migration and inbreeding. In the present study we have investigated the frequency of five polymorphisms within innate immune-response genes (SLC11A1 D543N, MBL2 G161A, P2RX7 A1513C, IL10 A-1082G, TLR2 –196 to –174 ins/del) related to susceptibility to infections in humans. The DNA of individuals from two early Roman-Period populations of Linowo and Rogowo was analysed. The distribution of three mutations varied significantly when compared to the modern Polish population. The TAFT analysis suggests that the decreased frequency of SLC11A1 D543N in modern Poles as compared to 2nd century Linowo samples is the result of non-stochastic mechanisms, such as purifying or balancing selection. The disparity in frequency of other mutations is most likely the result of genetic drift, an evolutionary force which is remarkably amplified in low-size groups. Together with the FST analysis, mtDNA haplotypes' distribution and deviation from the Hardy-Weinberg equilibrium, we suggest that the two populations were not interbreeding (despite the close proximity between them), but rather inbreeding, the results of which are particularly pronounced among Rogowo habitants.


Although no sound evidence of population differentiation was found when comparing the samples of Linowo and Rogowo, it is worth noticing that the distribution of mtDNA haplotypes between these two settlements differs remarkably. Apart from the two haplotypes (rCRS and 16126C) that occur in both studied groups, no other pattern of mtDNA SNPs is shared between them. The lack of reflection of these dissimilarities in the FST analysis is probably a result of the low-size group which is more exposed to result bias or low diversity of haplotypes among Rogowo individuals. All of the above allows to draw the theoretical conclusion that although these two settlements date back to the same period and are located within 55 km (or around 160 km along the Vistula River) of one another, they are genetically remote.

Lewandowska et al., The genetic profile of susceptibility to infectious diseases in Roman-Period populations from Central Poland, Infection, Genetics and Evolution, Volume 47, January 2017, Pages 1–8,

See also...

R1a-Z280 from Early Bronze Age Northern Poland

Wednesday, July 27, 2016

Lipka Tatars vs Balto-Slavs

Huge difference in this ADMIXTURE bar graph from the recent Pankratov et al. paper between Lipka Tatars from Belarus and nearby Balts and Slavs. The Lipka Tatars are almost identical to Volga Tatars despite residing in their current homeland for about 500 years. I'm guessing the fact that they're Sunni Muslims might have something to do with it.

Pankratov, V. et al. East Eurasian ancestry in the middle of Europe: genetic footprints of Steppe nomads in the genomes of Belarusian Lipka Tatars. Sci. Rep. 6, 30197; doi: 10.1038/srep30197 (2016).

Saturday, June 18, 2016

Poles in the new Human Origins dataset

Harvard's Human Origins dataset is being updated with 238 new samples, including 23 from Poland (15 from Poznan in western Poland and 8 from Lublin in eastern Poland). It should be available for download soon at the Reich Lab website here, although many of the new samples will only be accessible to people who sign a waiver. Below is a Principal Component Analysis (PCA) from Lazaridis et al. 2016 featuring the new samples. Interestingly, most of the Poles, probably those from Poznan, cluster with Sorbs from eastern Germany.


Lazaridis et al., The genetic structure of the world's first farmers, bioRxiv preprint, posted June 16, 2016, doi:

Sunday, May 1, 2016

Ancient Polish admixture in Denmark

This ESHG 2016 presentation about Danish population structure is sure to be interesting. I wonder how the authors were able to discern ancient Polish admixture from more recent Polish admixture? Keep in mind that lots of Poles settled in Denmark during the past 150 years or so. For instance, former Danish national team goalkeeper Peter Schmeichel is half Polish. And Caroline Wozniacki is totally Polish.

Abstract: Denmark’s genetic history has never been studied in detail. In this work, we analysed genetic and anthropometrical data from ~800 Danish students as part of an outreach activity promoting genomic literacy in secondary education. DNA analysis revealed remarkable homogeneity of the Danish population after discounting contributions from recent immigration. This homogeneity was reflected in PCA and AMOVA, but also in more sophisticated LD-based methods for estimating admixture. Notwithstanding Denmark’s homogeneity, we observed a clear signal of Polish admixture in the East of the country, coinciding with historical Polish settlements in the region before the Middle Ages. In addition, Denmark has a substantially smaller effective population size compared to Sweden and Norway, possibly reflecting further lack of strong population structure. None of these three Scandinavian countries seems to have suffered a depression due to the Black Death in the Middle Ages. Finally, we used the students’ genetic data to predict their adult height after training a novel prediction algorithm on public summary statistics from large GWAS. We validated our prediction using the students’ self-reported height and found that we could predict height with a remarkable ~64% accuracy.

Athanasiadis et al., Nationwide genomic study in Denmark reveals remarkable population homogeneity, ESHG EMPAG 2016 Presentation Abstract, P18.091C

Update 24/08/2016: The paper is now available at Genetics and open access. See here: On the remarkable genetic homogeneity of Denmark.

Monday, March 21, 2016

Scandinavians in medieval Poland

The genetic evidence presented in this paper is underwhelming; a single, low resolution mtDNA haplogroup I haplotype that appears to be of Scandinavian origin because it was also found in remains from Iron Age Denmark (sample B5 in Melchior et al. 2008). However, the authors' conclusions are also based on archaeological evidence, and they also match recent isotopic results (see here).

Abstract: Contemporary historical anthropology and classical archaeology are concerned not only with such fundamental issues as the origins of ancient human populations and migration routes, but also with the formation and development of inter-population relations and the mixing of gene pools as a result of inter-breeding between individuals representing different cultural units. The contribution of immigrants to the analysed autochthonous population and their effect on the gene pool of that population has proven difficult to evaluate with classical morphological methods. The burial of one individual in the studied Napole cemetery located in central Poland had the form of a chamber grave, which is typical of Scandinavian culture from that period. However, this fact cannot be interpreted as absolute proof that the individual (in the biological sense) was allochtonous. This gives rise to the question as to who was actually buried in that cemetery. The ancient DNA results indicate that one of the individuals had an mtDNA haplotype typical of Iron Age northern Europe, which suggests that he could have arrived from that area at a later period. This seems to indirectly confirm the claims of many anthropologists that the development of the early medieval Polish state was significantly and directly influenced by the Scandinavians.

Płoszaj T. et al., Analysis of medieval mtDNA from Napole cemetery provides new insights into the early history of Polish state, Ann Hum Biol. 2016 Mar 11:1-4., DOI:10.3109/03014460.2016.1151550

Tuesday, February 2, 2016

PCA of RISE595, RISE596 and RISE598

I wasn't confident enough to run these three ancient genomes in Principal Component Analyses (PCA) when they were first published last year. Their SNP counts were too low for comfort. But since then I've discovered a few things about PCA projection, so here goes:

RISE595 and RISE596 are samples from present-day Montenegro dated to the Late Bronze Age and Iron Age, respectively. I haven't been able to dig up a lot of information about their archaeological contexts, but both are from Kurgan-like Tumuli burials.

If their PCA placements are correct, then I'd say they more or less resemble the samples from Bronze Age Hungary (marked on the plot as Hungary_BA). What this suggests is that their DNA, like their burial style, is in large part of steppe origin.

RISE598 is a Late Bronze Age sample from a bog burial near the present-day Lithuanian/Polish border. This individual clusters close to Unetice samples from the Czech Republic, Germany and Poland, as well as present-day Poles, which, at least to me, makes good sense.

See also...

101 ancient Eurasian genomes (Allentoft et al. 2015)

Monday, December 21, 2015

Bronze Age steppe warrior affinity test

Reich Lab et al. are now targeting more than 1.2 million sites in the ancient samples that they genotype, including over one million from the Illumina genotype array. This is great because it means that I can now run analyses with these ancients on the present-day populations in my Illumina array-based dataset using as many as 500K SNPs, which makes things more accurate and interesting.

Below are a few sets of D-statistics of the form D(Outgroup,X)(Population1,Population2), featuring several ancient populations from the recent Allentoft et al., Haak et al. and Mathieson et al. papers. I'm basically testing how much closer X is to either Population1 or Population2.

The results look very solid considering the spatio-temporal and archaeological origins of the samples. However, the outcomes for the smaller Northern European nations perhaps look slightly inflated. This is probably due to their peculiar demographic histories and resulting relative genetic homogeneity. At this stage, I don't know how to test this and limit its effects, if indeed it is a problem.

Poltavka 0.4678
Yamnaya_Kalmykia 0.4667
Yamnaya_Samara 0.4659
Srubnaya 0.4583
Corded_Ware_Germany 0.4554

Full output here

Sintashta 0.4584
Poltavka 0.4571
Srubnaya 0.4566
Yamnaya_Samara 0.4556
Afanasievo 0.455

Full output here

Poltavka 0.4617
Yamnaya_Samara 0.4594
Afanasievo 0.4556
Yamnaya_Kalmykia 0.454
Srubnaya 0.4522

Full output here

Yamnaya_Samara 0.4658
Afanasievo 0.4616
Yamnaya_Kalmykia 0.4602
Corded_Ware_Germany 0.4555
Srubnaya 0.4551

Full output here

Srubnaya 0.4572
Poltavka 0.4568
Sintashta 0.4534
Yamnaya_Samara 0.4518
Corded_Ware_Germany 0.4514

Full output here

Poltavka 0.4555
Srubnaya 0.454
Andronovo 0.4537
Yamnaya_Samara 0.4536
Corded_Ware_Germany 0.4523

Full output here

Andronovo 0.4567
Srubnaya 0.4553
Poltavka 0.4526
Yamnaya_Kalmykia 0.4507
Unetice_EBA 0.4506

Full output here

Poltavka 0.4571
Yamnaya_Samara 0.4565
Afanasievo 0.454
Corded_Ware_Germany 0.4538
Unetice_EBA 0.4528

Full output here

Afanasievo 0.4644
Poltavka 0.464
Yamnaya_Samara 0.463
Srubnaya 0.4523
Andronovo 0.4513

Full output here

Poltavka 0.4669
Yamnaya_Kalmykia 0.4615
Afanasievo 0.4612
Srubnaya 0.4554
Corded_Ware_Germany 0.4542

Full output here

See also...

qpAdm tour of the Eneolithic/Bronze Age steppe

At least three genetically distinct Indo-European migrations into South Asia