QpAdm results using new Allentoft data

I have been  on holidays and now back behind the workbench.   Today I ran a qpAdm analysis using all possible modern European against the new data from Allentoft et al. 2015.  Unfortunately many northeast and southeast European groups turned out to be problematic and gave high chisq values indicating lacking admixtures. Although I found a fitting ancient group and a solution for northeastern groups it is still under test and at first I publish only results with chisq values below 1,5, only Estonian, Lithuanian and Basque being over 1.    Siberian indicates modern Siberian populations (Nganasan) and Near East indicated Bedouin-like admixture, used before also by Haak et al.

Bronze and Iron Age samples analyzed using Dstat

A month ago we saw a new study,  Allentoft et al. with new earlier unpublished data regarding several Bronze Age cultures.  Altogether 101 ancient samples were available, of which almost half has reasonable high quality.   I ran several PCA’s and noticed some problems due to the error caused by those low quality samples and obviously nonrandom distribution of SNP results.  If I used standard methods most new samples clustered somewhere between Central Europe and Caucasus and if I used the projection method included to Eigensoft’s PCA-tool most samples from ancient European cultures were placed among modern Europeans.  So I understood that PCA wouldn’t work well and wouldn’t reveal original ancient features and I saw it necessary to use straight comparisons between ancient and modern samples, comparing them without selective clustering.  Tools like f3Stat and Dstat are straightforward methods without for low quality data vulnerable clustering.  Therefore f3- and Dstat are more applicable in this case.

  

My first test includes selected Eastern European populations comparing them to other modern Europeans and ancient samples.  I used Dstat and the formula is Dstat(test-a,test-b;ancient sample,Mbuti), where “test-a” is the East European sample to be tested and “test-b” is the European sample to be compared with “test-a”.  If the result is positive then “test-a” is closer the ancient sample in comparison between "test-a" and "test-b".  If the result is negative then “test-b” is closer than "test-a".  I moved some results to Excel sheet to show one idea how to make comparisons.  New data is downloadable here. 

I publish now some first observations.  Although the locality seems to be absolutely right, ancient Scandinavian are close modern Scandinavians etc., there are many surprising results which are in contradiction with results obtained by selective clustering methods.  You are welcome to leave your comments if you find something surprising.  Unfortunately the publicly available version of Allentoft et al. doesn’t show comparable results using f3- or Dstat, so he keeps us in excitement. 

I have now only a few results from East Europe, but I’ll run more results including Central, West and South Europeans during the next week.

Examples click here.

edit 1.7.2015 11.40 am:  German samples are from Estonian BC and not representative.  They seem to be partly more unknown East Europeans than Germans from Germany. I should have deleted them.  

edit 4.7.2015 11.10 am:   More results, including Western Europeans, click here to download xlsx-sheet.

Albania and Sardinia, best remains of Early European Neolithic Era

I have opposed ideas calling Neolithic Europeans Near Easterners, because all my analyses show significant later Near Eastern admixture in Mediterranean region, clearly differing from ancient Neolithic farmer genes.  This implied at least equal genetic change in Near East than in Europe from the Neolithic Era to the present.  How can we call European Neolithic farmers Near Easterners if European Neolithic samples are closer modern European than modern Near Easterners?   Really, my analyses prove this.  Closest European for ancient Neolithic farmers are Albanians and Sardinians.  Albanian yDna E is one of the highest in Europe, after Kosovo, which can predict same thing.  Unfortunately Kosovo samples are not available.   It is very possible that European Neolithic farmers came from Near East through Anatolia, but we have not yet genetic evidences proving it and the process was probably much more complex than a simple migration to Europe from what we now understand as Near East.

My analysis consists of following steps:

1 pre-selection of  data (Haak et al. 2015)

2 running PCA to find out European Neolithic genetic cluster based on ancient samples of highest quality

3 running new PCA and Fst-figures to see genetic distance

This is a very simple procedure and I know well that Fst-distances obtained using ancient genomes are not comparable to distances between modern populations due to lower scanning quality, but  I don’t even try do it, I only compare ancient data to modern one, so possible quality problems doesn’t exist in same extent.

Please notice also many Jewish groups I included now into my data. 

PCA, click here to expand.

Fst-distances of Early Neolithic farmers, click here.

Fst-distances of Bedouins (BedouinB, used as a Near Eastern proxy by Haak), click here.

Three-way admixture of Central Europeans, Northeast and South show more admixtures

With respect to the Haak et al.  2015 and after my own pretests using qpAdm I ran a serie of standard admixture tests in supervised mode to find out ancient admixtures among Europeans.   Admixture gives a better way than qpAdm to use larger data sets, but without any quality indicators needs pretests if you want to run in supervised mode.  I also used some present-day populations to obtain a better total fit in cases previous tests and studies gave reason to expect such being reasonable.  In general my results are similar with those published by Haak.   K-group sizes differ, in some cases my results look more plausibe than corresponding results in the above mentioned study, for example my admixture shows less Siberian admixture in West Europe, but also more Near Eastern admixture in South Europe.   Those two admixtures came probably later to Europe which can be seen in their smaller expansion areas.

The first analysis (I'll do later more similar analyses) shows following k-groups:

1 WHG - Western Hunter-Gatherers
2 ENF . Early Neolithic Farmer (in Europe)
3 Yamnaya - samples from Yamnaya culture
4 East_Asia - Chinese She population
5 NA - Native Americans (Karitianas from South America)
6 SSA - Sub-Saharan African (Yoruba)
7 Near_East - Bedouins (BedouinB from Haak et al.)
8 Siberian - Nganassans

Sorted by ENF group averages.  Click here to see a large picture.

Finnish I1, can it descend from pre-Saxon ancestry, part 2

As I mentioned in my previos text, around 28% of Finnish men belong to a certain I1-clade named usually a Bothnian.  The rest 72% consists mostly of N1c1 clades and the rest, below 10% belong mostly to different R haplogroups.  The N1c1 splits roughly into two or three clades:  eastern Karelian, eastern Savonian clades and southern Baltic and Finno-Baltic clades.

The history of Finnish I1 was pretty long unkown and was a target of speculations although many Finns being aware of the Finnish history were able to see more and understand the course of history, actually pretty well (on our Finnish forums treating the Finish history was a member calling himself Moses Leone.  He stated 8 years ago that the Finnish I1 was older than usually supposed and bound to Iron Age migrations).   What we know now more is based on  new Y-chromosome related information.  It shows us hidden history of male ancestral lines thousands years back.  And what can we now say for sure based on the new information? It is the same as many hints have given for us to presume already years ago.  So we don't know now more basics, we only can broaden our knowledge with new facts.

Keeping in mind the Finnish history, not only genetics but also archaeological and linguistic evidences (known already around 200-300 years) it is undeniable that the Bothnian group is closely related to a certain Central European clade, known by the ydna-mutation CTS2208.   Out of touch of this mutation the relation can be confirmed also by STR-statistics.  STR-statistics can give misleading information if used to explain a single individual result more than hundreds years back in time, but statistical results are still valid.  The more we have high classified samples the more history can be plumbed.

I did the following test, you can do it too:

1.  Dowload all I1-L22 (L22 is an upstream mutation for CTS2208, L22>Z74>CTS2208, look here ) from FamilyTreeDna's global I1-project.  Data is available here
2.  Select only samples with 67 markers or more
3.  Select samples randomly from each L22-subclade to have almost equal amounts of samples in each subclade
4.  Run the data using some TMRCA calculator, you can do it for example here
5.  Run the generation table using some PCA- or MDS-software.  PAST works fine on Windows, downloadable here. You can also make phylogenetic trees, if you wish.

Here are my results.

Blue -    Bothnia, corresponds SNP mutation L258
Purple - "pre-Bothnia", mutation L287, negative for L258
Red -     "prepre-Bothnia", "pre-Saxon", mutation CTS2208, negative for L287
Yellow - "preprepre-Bothnia mutation Z74,  negative for CTS2208
Brown - L22-root, samples negative for all downstream mutations
Green -  combined Scandinavian L22-clades


PCA-plot:

   
Clusters k3, k4 and k5:

edit 30.4.15

According to the manager of Finnish I1-project the kit 38676 could be CTS2208+, predicted by STR-results and the kit has not yet tested CTS2208.  Assuming this being true we get following PCA-results.  I also added kit numbers and countries for all CTS2208 results.   If I dare to make some conclusions I would say that main Finnish I1-clades descend from a Scandinavian group which rose in Southern Scandinavia during the Bronze Age.  Evaluating more the history I would say that some men migrated from Southern Scandinavia to Finland around 2000 years ago and perhaps later same Scandinavians migrated to England.  English samples are overrepresented due to the US connection.

Finnish I1, can it descend from pre-Saxon ancestry?

It has gone almost year since I last time visited FtDna's I1-project, being quite tired of the usual pathetic ydna discussion. Now after a little pushed by another Finnish guy I took a look and noticed that the SNP-based phylogenetic tree shows more upstream matches for Finnish Bothnians than a year ago. Most of new matches are from England which obviously mirrors the testing activity more than enything else. Anyway, it looks like the Bothnian I1 descends from something like Anglo-Saxons, but we have to date it, accoring to the clade age (made by Ken Nordtvedt and others), to the time before Anglo-Saxon British migrations.  Here are those connected directly to Bothnians, They form an own class between all other I1 branches and Bothnians.

This information is public and available for anyone from FamilyTreeDna's project page and I don't see any reason to hide kit numbers and names.  I however overwrote all personal information because I am not sure about my right to publish Familytreedna's data.   Anyway it is public, here

https://www.familytreedna.com/public/yDNA_I1/default.aspx?section=yresults
  

 I1-L22 CTS2208 CTS5476 (L22+, Z74+, CTS2208+, CTS5476+
 XXXXXXXXXXXX Topsfield England
 XXXXXXXXXXXX SFK England
 XXXXXXXXXXXX Ryazan Russian Federation
 XXXXXXXXXXXX Glemminge Sweden
 XXXXXXXXXXXX England
 XXXXXXXXXXXX NY England
 XXXXXXXXXXXX United Kingdom
 XXXXXXXXXXXX RI England
 XXXXXXXXXXXX Luton Bedfordshire England
 XXXXXXXXXXXX Ephraim England
 XXXXXXXXXXXX New Jersey England
 XXXXXXXXXXXX England
 XXXXXXXXXXXX England
 XXXXXXXXXXXX Ephraim England
 XXXXXXXXXXXX England
 XXXXXXXXXXXX Germany
 XXXXXXXXXXXX Malderen, Belgium

And here is the beginning of the long Bothnian list. They represent about 28% of all Finnish men.

I1-L22 L258 (L22+, Z74+. CTS2208+, CTS5476+, L287+. L258+) [ Z133, BY2572, Y8598 available]
XXXXXXXXXXXX Mouhijärvi Finland
XXXXXXXXXXXX Pihlajavesi Finland
XXXXXXXXXXXX Kalajoki Finland
XXXXXXXXXXXX Finland
........

Comparing ancient and modern Europeans, a special case

Children of Native American and European parents share their parents’ autosomal dna.  Uniparental dna is however inherited without recombination.  The situation is more complex if we imagine an isolated population which started to grow after ancient migrations from 100 Native American men and 100 European women.  After several generations autosomal dna is more mixed. Autosomal dna is after generation also drifted, meaning they have lost something about the original information.  I think that genetic drift is like entropy.  The counterforce for entropy in this case is selection.  Selection resists genetic drift but only in chromosome regions where the selective impact is efficient, in other regions genetic drift goes on.   Uniparental dna doesn't transform similarly.  

What I now portrayed can be tested in practice, we need only two populations showing some common uniparental dna, being on autosomal side far enough from each other to give clear difference and also better if they are enough homogeneous proving isolation.  It is important that populations are not too much mixed. And I have them:  on the other side Sardinians (group 1) and then Finns and  Lithuanians (group 2).  Then we need to know who they were in the beginning. Because we can’t have ancient samples corresponding exactly to history of any modern populations I have used  EN results (European Early Neolithic samples), corresponding to the historical continuity of the group 1.  This historical similarity between Sardinians and EN was shown in Haak et al. 2015, being 91-96%.    

Here we go.  I gathered YDNA and MTDNA distributions from Eupedia and autosomal figures from  Haak et al. 2015.   If my theory is right we should see something like:

•  autosomal parity between ancient and modern populations in group 1 (Sardinians and EN) is high if both are mostly unmixed.  This is a postulate.

•  amounts between ancient autosomal dna (EN) and modern uniparental genes (group 2)  roughly match because uniparental genes don’t drift.

• amounts between modern autosomal genes (group 1, Sardinians) and modern autosomal genes (group 2) don’t match similarly as in previous comparison because autosomal genes have drifted in both populations, in this case around 7000 years, less or more hard to say exactly because we know too little about ancient migrations. 

But this is not a full story.   Trying to make accurate comparison between modern populations usually fails less or more, because we have no fixed reference points.   This happens espacially when using formal admixture softwares.  It happens also in this case when comparing Sardinians and Finns/Lithuanians without using EN as a reference.  Only ancient genomes can be reasonable references, so we actually don't know how much Sardinians and Finns/Lithuanians could match.   

YDNA and MTDA statistics gathered from Eupedia, to download click here.

Comparison between Haak et al. 2015, statistics from page 122 onwards (download here) ,  and now gathered Eupedia numbers:

Finns:  EN 31,5-42,5 %  corresponding to the modern common ydna uniparental with Sardinians 34,5 % (Eupedia numbers)

Lithuanians: EN 33,4-44,2 %  corresponding to the modern common ydna uniparental with Sardinians 38,65 % (Eupedia numbers)

We know that uniparental and autosomal genes correlate and it happens also here. 

For comparison three formal admixture analyses: