search this blog

Monday, September 18, 2017

Ancient IBD/cM matrix analysis offer


I've had a few requests from personal genomics customers to stick their files into an Identity-by-Descent/cM matrix like the one at the link below. Also please check out the accompanying comments thread for ideas of what can be done with the output.

A Bronze Age dominion from the Atlantic to the Altai

I can do this for $15 (USD) per individual. Please e-mail the data and money (via PayPal) to eurogenesblog [at] gmail [dot] com. The deadline for sending through the data files (which, in this run, can only be from 23andMe, Ancestry or FTDNA) is this time Tuesday.

I'll send out the results to each participant over e-mail. However, participants are encouraged to post their results in the comments thread below so that they can be discussed and analyzed further.

Update 20/09/2017: The analysis is underway. Please don't send any more data files. If there's enough interest, I'll do another run soon.

Update 22/09/2017: I've just sent out the results to the participants in the form of two text files titled "ancients_only" and "full_column". The former is a matrix of overall shared haplotype tracts in centimorgans (cM) that includes the user and 65 ancient genomes, and the latter a list of haplotype tracts, also in cM, shared between the user and well over 3000 public samples.

So what can we do with these files? For one, we can look at them, because simply eyeballing these sorts of stats can be very informative. Sorting the data in some way and calculating population averages might help with that.

The "ancients_only" file can be used for slightly more advanced analyses. For instance, below is a Neighbor joining graph produced with the Past 3 program (freely available here). I simply loaded my "ancients_only" file into Past 3, selected all of the columns and rows, and then did this: Multivariate > Clustering > Neighbor joining. Note that I cluster on the same branch as Slav_Bohemia, and this makes perfect sense considering my Polish ancestry. By the way, I dropped Oetzi from this run because he was behaving strangely, which is not unusual for low coverage genomes. Click on the image and open in a new tab for a better view.

Indeed, Past 3 can do a lot of interesting things with matrix files; anything from linear models to rotating three dimensional plots. If you'd like to repeat the linear models from my above linked to blog post, then choose the relevant two columns in your matrix and go Model > Generalized Linear Model. You should see something like this.


Moreover, a matrix with the 3000+ public samples can be gotten here and combined, in part or in whole, with your other files so that you can analyze yourself alongside a larger number of individuals.

54 comments:

Davidski said...

Should have these done on Thursday, unless something comes up.

Chad Rohlfsen said...

I'll have both to you in about 7-8 hours.

Davidski said...

Yeah, that's fine. I'm starting in about 24 hours.

Lathan The Great said...

Hey Davidski, when are the Tanzanian South-Cushitic pastoralist genomes being published?

And have you heard of any Neolithic and Paleolithic genomes from Northeast Africa yet? Especially the Nubia?

Thanks. And this is a great project.

epoch2013 said...

@David

Will you also check ancients? In other words, will you check if Iran Chalcolithic shares with Yamnaya rather than others? I mean, it may settle a huge number of debates, or at least give a hint at where to look.

Davidski said...

@Lathan

No idea when the African genomes are coming out. The paper is probably in peer review now.

@epoch

Iran_ChL genomes are pseudo-haploid. Can't run them in this.

epoch2013 said...

@David

O, I see. Not everything can be used for this.

Another thing, probably slightly OT. I came across this remark from Nick Patterson in the "Early Slavs from Bohemia" thread:

"This kind of thing is inevitable. There can be lab mixups of various types or reburials (so a body winds up in deeper strata). This is a reason that I am nervous about making heavy inference from samples that are outliers in a population.
a) Outlier might be of real interest
b) Or the sample has an atypical history
c) Or something is wrong.

Often hard to tell. The Reich lab is carbon dating more and more samples though this is expensive!"

http://eurogenes.blogspot.com/2017/05/two-early-slavs-from-bohemia.html?showComment=1495222388871#c5890778235093582920

From what I read in the Sup Info of Mathieson 2015 they did not mention carbon dating of the remains, only of the site. Do you know how sure we can be that the Khvalynsk outlier actually is a Khvalynsk sample?

Davidski said...

Do you know how sure we can be that the Khvalynsk outlier actually is a Khvalynsk sample?

He wasn't actually marked as an outlier in any papers to date.

But I don't know how his archaeological affiliation was determined exactly and whether it was a high confidence call. You might need to ask Nick Patterson about that.

Ryan said...

@David - But I don't know how his archaeological affiliation was determined exactly and whether it was a high confidence call. You might need to ask Nick Patterson about that.

You mean getting hit over the head and dumped into a ditch isn't specific to one culture? :3

I'm afraid I'll have to sit this out. Family health crisis unfortunately. Next time.

Davidski said...

@All

The analysis is underway. Please don't send any more data files. If there's enough interest, I'll do another run soon.

David Rabaez said...

Hi, this is my matrix. Now it remains to know its meaning. Comments!

https://drive.google.com/file/d/0B1HxVzk_qIwzd3pqbFFCSVFDSjg/view?usp=sharing

Thank you

David

Davidski said...

OK, I sent out the output files not long ago and updated the blog post with some ideas and instructions. Check out my Neighbor joining tree!

https://2.bp.blogspot.com/-ue-j4khjt_U/WcUcpe1vBeI/AAAAAAAAGGA/gH_4xNlXy_oBBr-6SeS9TP8W9hh8GYRQwCLcBGAs/s1600/David_W_Neighbor_joining_tree.png

Also, Matt, if you're reading this, here's the matrix with over 3,000 public samples.

https://drive.google.com/file/d/0B8XSV9HEoqpFOUVQd1lfejREYlk/view?usp=sharing

Helgenes50 said...

@ Davidski

Thanks for your work.
My results

AAA_Helgenes
AAB_Sintashta_RISE395 16.7517755
AAB_Roman_Britain_6DRIF-22 12.372527
AAB_Unetice_RISE577 11.72396279
AAB_Unetice_RISE150 11.3984053
AAB_Anglo_Saxon_NO3423 10.795745
AAB_Portugal_LNCA_DolmenAnsiao96B 10.53239
AAB_Afanasievo_RISE511 9.507238
AAB_Nordic_IA_RISE174 8.849702
AAB_Mezhovskaya_RISE523 8.82688
AAC_Karasuk_outlier_RISE502 8.078086
AAB_Yamnaya_Kalmykia_RISE552 6.857887
AAB_Spain_LNCA_ATP2 5.91294
AAB_Yamnaya_Kalmykia_RISE548 5.791653
AAB_Nordic_LN_RISE98 5.679253
AAC_Altai_IA_RISE602 5.554495
AAB_Portugal_MBA_TV3831 5.52035
AAB_Spain_LNCA_Matojo 5.059983
AAB_Ireland_EBA_Rathlin2_RSK1 4.916634
AAB_Portugal_MN_LugarCanto42 4.45519
AAB_Ireland_EBA_Rathlin1_RM127 4.422253
AAB_Hungary_IA_IR1 4.419994
AAB_Slav_Bohemia_RISE569 4.37205
AAB_Hungary_HG_I1507 4.371752
AAC_Karasuk_outlier_RISE495 3.959117
AAB_Ireland_MN_Ballynahatty_BA64 3.826798
AAC_Karasuk_outlier_RISE493 3.643144
AAB_Satsurblia_SATP 3.613877
AAB_Portugal_MBA_TV32032 3.594428
AAB_Roman_Britain_6DRIF-3 3.533179
AAB_Hungary_BA_RISE479 3.44839
AAB_Greece_LN_Pal7 3.179656
AAB_Roman_Britain_6DRIF-18 3.03813
AAB_Hungary_N_I1506 2.985472
AAB_Iberia_EN_CB13 2.903345
AAC_Altai_IA_RISE601 2.63249
AAB_Karasuk_RISE496 2.53847
AAB_Motala_HG_Motala12 2.36684
AAB_Anatolia_N_Bar31 2.350238
AAB_LBK_Stuttgart 1.977932
AAB_Sweden_MN_Gok2 1.85787
AAB_Karasuk_RISE499 1.84502
AAB_Bichon_Bichon 1.84386
AAB_Portugal_MN_LugarCanto41 1.766666
AAB_Greece_LN_Klei10 1.67404
AAB_Portugal_LNCA_CovaMoura9B 1.63677
AAB_Portugal_MBA_MonteGato104 1.557905
AAB_Hungary_N_I1495 1.5414
AAB_Greece_EN_Rev5 1.52268
AAB_Roman_Britain_6DRIF-21 1.486628
AAB_Andronovo_RISE500 1.11802
AAB_Hungary_BA_I1504 0.85747
AAB_Anatolia_N_Bar8 0
AAB_Andronovo_RISE505 0
AAB_Hungary_CA_I1497 0
AAB_Hungary_N_I1496 0
AAB_Kotias_KK1 0
AAB_LaBrana1_LaBrana1 0
AAB_Loschbour_Loschbour 0
AAB_Oetzi_Oetzi 0
AAB_Pitted_Ware_Ajv58 0
AAB_Portugal_LNCA_CabecoArruda122A 0
AAB_Portugal_LNCA_CovaMoura364 0
AAB_Portugal_MN_LugarCanto44 0
AAB_Roman_Britain_outlier_3DRIF-26 0
AAB_Spain_LNCA_ATP16 0
AAC_Altai_IA_RISE504 0
AAC_Karasuk_outlier_RISE497 0







Derek said...

I'm American, 7/8th Norwegian by descent.

Top modern matches, averaged:


RussiaNW 10.3628055
Norwegian 8.5535599
Utah_Scandinav 8.292284163
Saami 7.2426796
Finnish 6.643150417
Utah_German 6.054027664
Orcadian 5.65797413
Russian_North 5.088347215
English_Kent 5.083618765
Estonian 4.835378524
Cornwall 4.822076878
Utah_USA 4.469781414

Only two Russian Northwest and three Saami samples, but there seems to be a real connection. I score as about 2% Siberian on ADMIXTURE calculators, so it makes sense. Two of my three closest individual matches were to Chuvash.

Ancient individual matches:

AAB_Unetice_RISE150 13.442742
AAB_Nordic_LN_RISE98 12.851084
AAB_Yamnaya_Kalmykia_RISE548 11.005777
AAC_Karasuk_outlier_ 9.004888
AAB_Karasuk_RISE499 8.809091
AAB_Roman_Britain_6DRIF 8.637628
AAB_Ireland_EBA_Rathlin 8.606186
AAB_Andronovo_RISE505 8.247044
AAB_Ireland_EBA_Rathlin2 _ 7.941592
AAB_Andronovo_RISE500 7.14226013
AAB_Greece_LN_Pal7 6.834963


Small samples, so I don't know whether to make a big deal out of these results. My closest matches are to an Unetice Pole from around 1800 B.C. and an R1b Swede from around 2100 B.C. The other Unetice sample was a dud, so I suspect that with enough data I would turn out closest to the Neolithic Scandinavians.

Paul Brooker said...

English, East Anglian:

AAB_Afanasievo_RISE511 9.232964
AAB_Anatolia_N_Bar31 0
AAB_Anatolia_N_Bar8 1.973855
AAB_Andronovo_RISE500 10.2060098
AAB_Andronovo_RISE505 0
AAB_Anglo_Saxon_NO3423 6.5713445
AAB_Bichon_Bichon 0.7067
AAB_Greece_EN_Rev5 0
AAB_Greece_LN_Klei10 0
AAB_Greece_LN_Pal7 1.76034
AAB_Hungary_BA_I1504 2.36881
AAB_Hungary_BA_RISE479 0
AAB_Hungary_CA_I1497 1.625764
AAB_Hungary_HG_I1507 0.96928
AAB_Hungary_IA_IR1 3.5563518
AAB_Hungary_N_I1495 4.272295
AAB_Hungary_N_I1496 0
AAB_Hungary_N_I1506 2.866862
AAB_Iberia_EN_CB13 2.802699
AAB_Ireland_EBA_Rathlin1_RM127 8.597182
AAB_Ireland_EBA_Rathlin2_RSK1 12.52403
AAB_Ireland_MN_Ballynahatty_BA64 3.387557
AAB_Karasuk_RISE496 3.698784
AAB_Karasuk_RISE499 20.6952997
AAB_Kotias_KK1 0
AAB_LaBrana1_LaBrana1 1.4578
AAB_LBK_Stuttgart 2.12978
AAB_Loschbour_Loschbour 2.608912
AAB_Mezhovskaya_RISE523 1.075
AAB_Motala_HG_Motala12 3.683518
AAB_Nordic_IA_RISE174 7.2245
AAB_Nordic_LN_RISE98 3.9755169
AAB_Oetzi_Oetzi 1.97706
AAB_Pitted_Ware_Ajv58 2.346145
AAB_Portugal_LNCA_CabecoArruda122A 10.801202
AAB_Portugal_LNCA_CovaMoura364 0
AAB_Portugal_LNCA_CovaMoura9B 2.56866
AAB_Portugal_LNCA_DolmenAnsiao96B 0
AAB_Portugal_MBA_MonteGato104 1.462675
AAB_Portugal_MBA_TV32032 0
AAB_Portugal_MBA_TV3831 4.378157
AAB_Portugal_MN_LugarCanto41 2.63479973
AAB_Portugal_MN_LugarCanto42 6.89848067
AAB_Portugal_MN_LugarCanto44 1.71684
AAB_Roman_Britain_6DRIF-18 5.6299293
AAB_Roman_Britain_6DRIF-21 1.767712
AAB_Roman_Britain_6DRIF-22 4.019409
AAB_Roman_Britain_6DRIF-3 3.682746
AAB_Roman_Britain_outlier_3DRIF-26 1.15846
AAB_Satsurblia_SATP 0
AAB_Sintashta_RISE395 2.696899
AAB_Slav_Bohemia_RISE569 1.920134
AAB_Spain_LNCA_ATP16 4.232946
AAB_Spain_LNCA_ATP2 1.889402
AAB_Spain_LNCA_Matojo 3.6635848
AAB_Sweden_MN_Gok2 0
AAB_Unetice_RISE150 6.551865
AAB_Unetice_RISE577 2.46105
AAB_Yamnaya_Kalmykia_RISE548 8.7826224
AAB_Yamnaya_Kalmykia_RISE552 1.41141
AAC_Altai_IA_RISE504 2.04188
AAC_Altai_IA_RISE601 0
AAC_Altai_IA_RISE602 0
AAC_Karasuk_outlier_RISE493 4.347327
AAC_Karasuk_outlier_RISE495 1.184342
AAC_Karasuk_outlier_RISE497 0
AAC_Karasuk_outlier_RISE502 1.737628

Matt said...

@ Davidski, cheers for the shoutout. The size of that matrix is really at or beyond the limits of what I can do with averaging out with the spreadsheet method I use (or at least, it maybe takes longer than i have patience for ;) ) but thanks.

Chad Rohlfsen said...

@ Matt,

I'm sure everyone would be okay with you using our data for graphing. You're free to use mine, for sure. I'll have my stuff up here soon.

Chad Rohlfsen said...

Euro-mutt

Ancients:

AAB_Nordic_LN_RISE98 15.1803537
AAC_Karasuk_outlier_RISE502 12.14899901
AAB_Yamnaya_Kalmykia_RISE548 11.939083
AAB_Afanasievo_RISE511 11.718604
AAB_Unetice_RISE150 10.876749
AAB_Andronovo_RISE505 10.424346
AAB_Karasuk_RISE499 9.91026
AAB_Yamnaya_Kalmykia_RISE552 9.335215
AAB_Sintashta_RISE395 7.3357957
AAB_Anglo_Saxon_NO3423 7.138464
AAB_Ireland_EBA_Rathlin1_RM127 6.914043
AAB_Roman_Britain_6DRIF-22 6.540072
AAB_Ireland_EBA_Rathlin2_RSK1 6.033046
AAB_Andronovo_RISE500 5.982405
AAB_Portugal_MBA_TV32032 5.855381
AAB_Portugal_MN_LugarCanto42 5.715064
AAB_Sweden_MN_Gok2 5.332439
AAB_Hungary_CA_I1497 5.1534879
AAB_Hungary_BA_I1504 5.109243
AAB_Roman_Britain_6DRIF-3 5.09514
AAC_Altai_IA_RISE504 4.871235
AAB_Pitted_Ware_Ajv58 4.278134
AAB_Portugal_MBA_MonteGato104 4.19162144
AAB_Spain_LNCA_Matojo 4.089124
AAB_Karasuk_RISE496 3.6737917
AAC_Altai_IA_RISE601 3.475469
AAC_Karasuk_outlier_RISE493 3.366392
AAB_Unetice_RISE577 3.302024
AAB_Nordic_IA_RISE174 3.085825
AAB_Mezhovskaya_RISE523 3.060812
AAB_Ireland_MN_Ballynahatty_BA64 3.020314
AAB_Portugal_MN_LugarCanto41 2.9059
AAB_Roman_Britain_6DRIF-18 2.860906
AAB_Loschbour_Loschbour 2.730059
AAB_Spain_LNCA_ATP16 2.66654813
AAB_LBK_Stuttgart 2.572323
AAB_Hungary_N_I1495 2.54884
AAC_Altai_IA_RISE602 2.4768
AAB_Greece_EN_Rev5 2.354084
AAB_Motala_HG_Motala12 2.008695
AAB_Hungary_HG_I1507 1.899536
AAB_Portugal_MN_LugarCanto44 1.856754
AAB_Hungary_BA_RISE479 1.837311
AAB_Greece_LN_Pal7 1.82286
AAB_Bichon_Bichon 1.759274
AAB_Roman_Britain_6DRIF-21 1.58591
AAB_Greece_LN_Klei10 1.52577
AAB_Spain_LNCA_ATP2 1.39874
AAB_Slav_Bohemia_RISE569 1.378009
AAB_Satsurblia_SATP 1.282883
AAB_Portugal_LNCA_CovaMoura364 1.266809
AAB_Hungary_IA_IR1 1.06789
AAB_Portugal_LNCA_CabecoArruda122A 1.056011
AAB_Portugal_LNCA_CovaMoura9B 0.986935
AAB_Oetzi_Oetzi 0.75436
AAB_Portugal_LNCA_DolmenAnsiao96B 0.580779
AAB_Anatolia_N_Bar31 0.28784
AAB_Anatolia_N_Bar8 0
AAB_Hungary_N_I1496 0
AAB_Hungary_N_I1506 0
AAB_Iberia_EN_CB13 0
AAB_Kotias_KK1 0
AAB_LaBrana1_LaBrana1 0
AAB_Portugal_MBA_TV3831 0
AAB_Roman_Britain_outlier_3DRIF-26 0
AAC_Karasuk_outlier_RISE495 0
AAC_Karasuk_outlier_RISE497 0

Chad Rohlfsen said...

Top 100 modern individuals:

Utah_USA_NA11891 23.4682238
Latvian_GS000035027 19.5256044
English_Cornwall_HG00249 16.433908
Utah_Scandinavian_NA07346 14.635498
Utah_German_NA12239 14.4500472
Utah_German_NA07022 14.125439
Mari_mari3 12.778314
Russian_North_HGDP00891 12.629419
Orcadian_HG00125 12.290583
Orcadian_HGDP00800 11.930816
Orcadian_HG00124 11.640142
Utah_USA_NA12145 11.291805
Orcadian_HGDP00804 11.195737
English_Cornwall_HG00263 10.948266
English_Kent_HG01334 10.86301
Norwegian_NOR101 10.77167783
French_HGDP00538 10.7579804
Belarusian_GS000014324 10.299699
Estonian_ee68 10.293421
Slovakian_Slovakia222 10.2055053
French_HGDP00528 10.150875
English_Kent_HG00128 10.087612
Serbian_Serbia_Serbia19_GSM1424665 10.058413
Italian_Tuscan_NA20509 9.891118
Spanish_Castilla_Y_Leon_HG01506 9.807251
Estonian_ee86 9.7524262
Utah_USA_NA12154 9.624001
Estonian_ee144 9.438562
Orcadian_HG00100 9.334663
Moksha_mordovia5 9.269123
Sicilian_West_WestSicilian10H 9.221184
Ukrainian_West_GS000035178 9.1698148
Finnish_HG00364 9.135117
Ukrainian_East_UkrBel618 9.036228
Ashkenazi_Jew_ashkenazy1e 9.012637
Orcadian_HG00117 8.919005
Ashkenazi_Jew_ashkenazy7e 8.86112
French_HGDP00513 8.760578
Komi_GS000035018 8.7295794
English_Cornwall_HG00256 8.669854
Spanish_Baleares_HG01670 8.6602
Utah_USA_NA12283 8.444657
Slovakian_Slovakia118 8.415299
Croatian_GSM1841131 8.410251
Utah_USA_NA12749 8.3990729
Utah_Scandinavian_NA12249 8.379915
English_Cornwall_HG00259 8.353
Russian_North_HGDP00887 8.286039
Utah_USA_NA12044 8.284005
English_Cornwall_HG00250 8.231026
Utah_USA_NA12282 8.21636
Norwegian_NOR126 8.200352
Orcadian_HG00102 8.165127
English_Kent_HG00141 8.156837
Utah_USA_NA12546 8.077786
Bulgarian_Bulgaria5 8.020719
Latvian_GS000016903 7.947704
English_Kent_HG00146 7.937961
Bosnian_Bosnian_4_GSM1424660 7.884202
Vepsa_GS000035244 7.878328
Spanish_Spain10 7.859405
Polish_Polish16H 7.804417
Slovenian_Slovenian_14 7.77683557
English_Cornwall_HG00235 7.740439
Serbian_Serbian_Serbia9_GSM1424683 7.701269
Romanian_Romania6 7.686074
English_Cornwall_HG00245 7.6763593
Utah_USA_NA12342 7.547523
Spanish_Baleares_HG01613 7.53747269
French_HGDP00531 7.5274779
Sephardic_Jew_sephardic5tur 7.498313
Macedonian_Macedonian2_GSM1424618 7.488779
Spanish_Galicia_HG01705 7.4170548
Lithuanian_lithuania3 7.410625
Estonian_ee66 7.381807
English_Cornwall_HG00234 7.331173
English_Cornwall_HG00264 7.309364
Utah_USA_NA12748 7.299198
Basque_French_HGDP01367 7.28241
Utah_USA_NA12045 7.240136
Estonian_ee45 7.1686808
French_HGDP00525 7.165268
Spanish_Castilla_La_Mancha_HG01746 7.1598775
Utah_USA_NA12340 7.148583
Estonian_ee50 7.140646
English_Kent_HG00157 7.11687
Lithuanian_lithuania6 7.103729
Kosovar_Kosovo12_GSM1424631 7.099791
Utah_British_NA12005 7.078816
Hazara_HGDP00112 7.072089
Estonian_ee136 7.047731
Ashkenazi_Jew_ashkenazy9e 7.042993
Slovenian_Slovenian188 6.999963
Orcadian_HG00108 6.995197
Norwegian_NOR107 6.977914
Basque_French_HGDP01379 6.908994
Russian_North_HGDP00901 6.899245
Latvian_latvian54H7 6.8962537
Croatian_GSM1841115 6.885017
Ukrainian_West_GS000035175 6.8425


Chad Rohlfsen said...

Top 75 modern pop averages:

Komi 6.4815982
Tatar_Mishar 6.227453
Utah_Scandinavian 6.18233262
Utah_German 5.81818997
Latvian 5.78009412
Utah_USA 5.5298965
Norwegian 5.29431744
Polish 5.21782346
Orcadian 5.08637953
Ingrian 4.93772933
English_Cornwall 4.81405562
Estonian 4.28759346
English_Kent 4.2115799
Ukrainian_West 4.13614942
Ukrainian_East 4.113382
Spanish_Baleares 4.0551462
Utah_British 4.03981567
French 3.95409879
Finnish 3.84986767
Vepsa 3.76677485
Belarusian 3.70539261
Spanish_Castilla_La_Mancha 3.68616588
Swedish 3.6024675
Slovenian 3.58938705
Slovakian 3.53558898
Ashkenazi_Jew 3.45085688
Spanish_Canarias 3.24010165
Russian_North 3.22359826
Hungarian 3.20907621
Moksha 3.1870786
Saami 3.18459067
Tatar_Kryashen 3.1266269
Lithuanian 3.03784985
Serbian 3.03708617
Romanian 2.96786836
Bosnian 2.8723482
Tabasaran 2.77484367
Russian_Northwest 2.759969
Mari 2.74862285
Montenegrin 2.66877436
Ukrainian_North 2.659621
Russian_West 2.628892
Croatian 2.62389354
Azeri 2.6129104
Basque_French 2.61282453
German 2.57363567
Spanish_Valencia 2.56562889
Spanish_Castilla_Y_Leon 2.51982362
Spanish_Cataluna 2.5122776
Mansi 2.49427491
Spanish_Cantabria 2.47494925
Spanish_Aragon 2.40970858
Macedonian 2.40772307
Erzya_mordovia 2.34740533
Spanish 2.27055123
Udmurd 2.23056375
Chuvash 2.20894819
Spanish_Andalucia 2.1904231
Kosovar 2.12828778
Greenlander_West 2.0854406
Bashkir 2.03821524
Kabardin 1.99343567
Spanish_Murcia 1.97721267
Italian_Tuscan 1.94721122
Hazara_Afghanistan 1.90979333
Cossack_Kuban 1.8445
Greek_Macedonia 1.79308271
Scottish_Argyll 1.73305925
Bulgarian 1.72939025
Spanish_pais_Vasco 1.6621105
Spanish_Extremadura 1.6221965
Spanish_Galicia 1.6128731
Italian_Bergamo 1.60662101
Cossack_Ukrainian 1.5517655
Moroccan_Jew 1.5394192

Dmytro said...

West Ukrainian culturally for 250 years paper trail.

"Ancients" above 5.0
AAB_Ireland_EBA_Rathlin1_RM127 13.1556885
AAB_Hungary_IA_IR1 9.87608
AAB_Unetice_RISE150 8.919633
AAB_Unetice_RISE577 7.762798
AAB_Andronovo_RISE505 7.653074
AAB_Ireland_EBA_Rathlin2_RSK1 7.16609
AAB_Hungary_HG_I1507 7.023542
AAB_Nordic_IA_RISE174 6.709605
AAB_Afanasievo_RISE511 6.353592
AAC_Karasuk_outlier_RISE502 6.105593
AAB_Hungary_BA_RISE479 5.91144
AAB_Portugal_MBA_TV32032 5.860841

"Moderns" above 16.0
Croatian_GSM1841121 22.940375
Slovakian_Slovakia411 21.47469428
Estonian_ee147 20.3318413
Estonian_GS000017209 19.5201645
Russian_North_HGDP00885 19.340754
Estonian_ee78 19.247257
Polish_Polish10H 19.190638
Lithuanian_lithuania9 18.685255
Ukrainian_West_UkrLv240 18.537329
Ukrainian_North_Ukraine136 18.468432
Estonian_a39 18.309442
Belarusian_GS000014324 18.236673
Ukrainian_West_GS000035178 18.127011
Ukrainian_West_UkrLv215 18.001162
Latvian_latvian54H7 17.9410166
Mozabite_HGDP01273 17.905884
Ukrainian_East_UkrBel736 17.866116
Slovakian_Slovakia256 17.604307
Belarusian_belorus7 17.385669
Polish_Polish12H 17.3128468
Ukrainian_North_UkrainePol19 16.905952
Ukrainian_East_UkrBel733 16.8683653
Russian_West_GS000016819 16.845141
Russian_North_HGDP00882 16.8322631
Latvian_latvian22J5 16.716382
Karelian_GS000035149 16.4635504
Polish_Polish9H 16.111341

Helgenes50 said...

Now top Moderns

Basque_French_HGDP01380 17.721878
Spanish_Valencia_HG01776 14.859439
English_Cornwall_HG00160 13.506335
English_Kent_HG00152 13.192521
Utah_Scandinavian_NA07346 12.043676
French_HGDP00519 11.449055
Estonian_GS000017199 10.875242
Utah_USA_NA12145 10.681289
English_Cornwall_HG00245 10.6283624
Vepsa_GS000016971 10.563117
French_HGDP00537 10.332528
Orcadian_HG00102 10.044597
Orcadian_HG00121 9.9740582
Polish_Polish11H 9.893548
Lithuanian_GS000016904 9.841885
English_Kent_HG00145 9.703207
Russian_North_HGDP00896 9.440439
Slovakian_Slovakia222 9.416012
Russian_North_HGDP00886 9.3189804
English_Cornwall_HG00263 9.106229
Orcadian_HG00107 9.033685
Moksha_mordovia14 9.0278884
Utah_Scandinavian_NA12248 8.94535
Italian_Tuscan_NA20766 8.92634
Utah_USA_NA12749 8.847244
Avar_GS000014407 8.806463
Bosnian_Bosnian_13_GSM1424654 8.702346
French_HGDP00535 8.70054022
Italian_Tuscan_HGDP01162 8.659078
Estonian_ee78 8.628335
Greenlander_East_GSM558768 8.366382
Norwegian_NOR106 8.344245
Utah_USA_NA12341 8.334125
Utah_USA_NA11893 8.286903
Sardinian_HGDP00673 8.2761015
Utah_USA_NA12264 8.272402
Spanish_Castilla_Y_Leon_HG01612 8.26236
Russian_Central_GS000035242 8.207584
Spanish_Baleares_HG01631 8.183449
Hungarian_GS000016901 8.168986
Scottish_Argyll_HG00104 8.098118
Utah_German_NA07345 8.087358
Utah_British_NA11920 8.028522
Italian_Tuscan_NA20813 8.00725
Serbian_Serbian_Serbia7_GSM1424682 7.9656198
French_HGDP00517 7.9191949
Lithuanian_lithuania6 7.909469
English_Cornwall_HG00259 7.80104
Orcadian_HG00114 7.751608
English_Kent_HG00130 7.649591
English_Kent_HG01334 7.616478
Estonian_GS000017206 7.512956
Colombian_GSM531114 7.50871
Utah_USA_NA12342 7.504915
Latvian_latvian58C6 7.502672
Italian_Tuscan_NA20527 7.455373
Utah_USA_NA12282 7.421038
Russian_North_HGDP00903 7.397888
Gujarati_NA20904 7.394681
French_HGDP00529 7.3924
Norwegian_NOR111 7.283133
English_Kent_HG00143 7.275521
French_HGDP00511 7.274362
Slovakian_Slovakia425 7.261354
Romanian_Romania12 7.261317
Orcadian_HGDP00803 7.243577
Utah_German_NA11830 7.230943
Orcadian_HG00109 7.22228
English_Kent_HG00129 7.193687
Utah_USA_NA12056 7.146905
Basque_French_HGDP01378 7.088122
Spanish_Spanish20H 7.018347
Estonian_GS000017208 6.989605
Macedonian_Macedonian7_GSM1424624 6.987235
Serbian_Serbian_Serbia5_GSM1424679 6.943076
Utah_USA_NA11881 6.931813
Sephardic_Jew_sephardic14bul 6.915986
Bulgarian_Bulgarian10H 6.855414
Greenlander_West_GSM558777 6.847348
English_Cornwall_HG00256 6.83909
Romanian_Romania4 6.827662
Russian_West_GS000016796 6.781069
Utah_USA_NA11919 6.759878
Spanish_Galicia_HG01707 6.751144
English_Kent_HG00127 6.747961
Ashkenazi_Jew_ashkenazy10e 6.738469
Lithuanian_GS000016905 6.723337
Latvian_latvian58C8 6.703595

David Rabaez said...

Ancients:

AAB_Spain_LNCA_ATP2 12.154296
AAB_Unetice_RISE150 9.608421
AAB_Anglo_Saxon_NO3423 9.3617289
AAB_Spain_LNCA_Matojo 8.606585
AAB_Nordic_LN_RISE98 7.6308697
AAB_Spain_LNCA_ATP16 7.18066
AAB_Portugal_MN_LugarCanto44 6.754815
AAB_Portugal_MBA_MonteGato104 6.68850484
AAB_Portugal_MBA_TV32032 6.5379326
AAB_Portugal_LNCA_CovaMoura364 6.4564237
AAB_LaBrana1_LaBrana1 6.397674
AAB_Ireland_MN_Ballynahatty_BA64 6.161949
AAB_Ireland_EBA_Rathlin2_RSK1 5.90511
AAB_Nordic_IA_RISE174 5.69514
AAB_Andronovo_RISE500 5.566669
AAB_Anatolia_N_Bar8 5.382943
AAB_Roman_Britain_6DRIF-21 5.336766
AAB_Portugal_LNCA_CovaMoura9B 5.151194
AAB_Andronovo_RISE505 4.884776
AAB_Bichon_Bichon 4.807908
AAB_Roman_Britain_6DRIF-18 3.87896
AAB_Portugal_MN_LugarCanto42 3.523487
AAB_Portugal_MN_LugarCanto41 3.516834
AAC_Karasuk_outlier_RISE502 3.464467
AAB_Yamnaya_Kalmykia_RISE552 3.452883
AAB_Mezhovskaya_RISE523 3.275702
AAB_Loschbour_Loschbour 3.243495
AAB_Slav_Bohemia_RISE569 3.219847
AAB_Motala_HG_Motala12 3.217487
AAB_Hungary_CA_I1497 3.188648
AAB_LBK_Stuttgart 2.9066
AAB_Ireland_EBA_Rathlin1_RM127 2.805886
AAC_Karasuk_outlier_RISE495 2.775052
AAB_Hungary_N_I1496 2.510278
AAB_Hungary_BA_I1504 2.44784
AAB_Iberia_EN_CB13 2.417861
AAC_Altai_IA_RISE601 2.407046
AAC_Karasuk_outlier_RISE493 2.368364
AAB_Yamnaya_Kalmykia_RISE548 2.339761
AAB_Greece_EN_Rev5 2.28117
AAB_Karasuk_RISE496 2.276359
AAB_Hungary_N_I1506 2.17003
AAB_Oetzi_Oetzi 2.157042
AAB_Roman_Britain_6DRIF-22 2.140597
AAB_Anatolia_N_Bar31 2.082461
AAB_Hungary_N_I1495 2.02615
AAC_Altai_IA_RISE602 1.9552
AAB_Afanasievo_RISE511 1.949347
AAB_Karasuk_RISE499 1.933334
AAB_Unetice_RISE577 1.9324181
AAB_Hungary_IA_IR1 1.91747
AAB_Portugal_LNCA_CabecoArruda122A 1.74305
AAB_Portugal_MBA_TV3831 1.73546
AAB_Portugal_LNCA_DolmenAnsiao96B 1.711982
AAB_Roman_Britain_6DRIF-3 1.585151
AAB_Hungary_BA_RISE479 1.36846
AAB_Greece_LN_Klei10 1.32308
AAB_Greece_LN_Pal7 1.21711

Top 40 moderns:

Basque_French_HGDP01363 9.88102
Spanish_Valencia_HG01605 9.8416581
Moroccan_Moroccan12H 9.6163749
Spanish_Aragon_HG01676 9.53385
Basque_French_HGDP01373 9.514234
Spanish_Extremadura_HG01510 9.429952
Spanish_Cataluna_HG01536 9.247764
Ukrainian_East_GS000035174 8.9539541
Utah_USA_NA12413 8.939456
Spanish_Cataluna_HG01537 8.69123
English_Cornwall_HG00237 8.655676
Basque_French_HGDP01375 8.651864
Montenegrin_Montenegro7_GSM1424640 8.637771
Mozabite_HGDP01259 8.494754
Peruvian_HG02302 8.485347
Basque_French_HGDP01367 8.45435
Spanish_Spanish20H 8.332209
German_GS000016891 8.297376
Puerto_Rican_GSM531210 8.0924
Spanish_Spanish24H 8.070807
Basque_French_HGDP01371 7.988825
Italian_South_SouthItalian4H 7.9041092
Jordanian_Jordan646 7.819715
Moroccan_Moroccan14H 7.810098
Estonian_ee68 7.749795
Basque_French_HGDP01365 7.728534
English_Cornwall_HG00259 7.58315
Scottish_Argyll_HG00103 7.500374
Italian_Tuscan_HGDP01168 7.4880377
Utah_Scandinavian_NA07346 7.433837
Spanish_Murcia_HG01617 7.396799
Colombian_GSM531127 7.306737
Spanish_Castilla_La_Mancha_HG01504 7.2744332
Karelian_GS000016970 7.2304023
Basque_French_HGDP01377 7.1713424
Tunisian_Tunisian8H 7.136262
Spanish_Valencia_HG01779 7.076908
Hungarian_hungary20 7.041261
Basque_French_HGDP01364 7.021052
Spanish_Castilla_Y_Leon_HG01783 7.018113

David Rabaez said...

Graphics:

- Oetzi - Rabai

https://drive.google.com/file/d/0B1HxVzk_qIwzZU1IVEVaVkRzS0E/view?usp=sharing

- Tree (Neighbour_joining_clustering)

https://drive.google.com/file/d/0B1HxVzk_qIwzWjE4STQwRzJ3WGs/view?usp=sharing

Thanks for you work, David!

Regards

Tesmos said...

My results: (Ancients)

AAB_Andronovo_RISE500 17,145621
AAB_Roman_Britain_6DRIF-22 15,995595
AAB_Yamnaya_Kalmykia_RISE548 11,343428
AAB_Afanasievo_RISE511 10,296394
AAB_Ireland_EBA_Rathlin2_RSK1 8,8230396
AAB_Nordic_IA_RISE174 8,101785
AAB_Andronovo_RISE505 7,780931
AAB_Roman_Britain_6DRIF-3 6,933925
AAB_Pitted_Ware_Ajv58 6,68318
AAB_Mezhovskaya_RISE523 6,510341
AAB_Roman_Britain_6DRIF-18 6,508469
AAC_Karasuk_outlier_RISE502 6,016296
AAB_Slav_Bohemia_RISE569 5,785742
AAB_Yamnaya_Kalmykia_RISE552 5,752832
AAB_Portugal_MN_LugarCanto41 5,44168
AAB_Nordic_LN_RISE98 4,919167
AAB_Spain_LNCA_ATP2 4,674716
AAB_Ireland_EBA_Rathlin1_RM127 4,477305
AAB_Portugal_LNCA_CabecoArruda122A 4,3234563
AAB_Iberia_EN_CB13 4,3009722
AAB_Portugal_LNCA_CovaMoura9B 3,934942
AAB_Hungary_N_I1506 3,876037
AAB_Unetice_RISE577 3,814262
AAB_Sintashta_RISE395 3,734422
AAB_Portugal_MBA_TV32032 3,591087
AAB_Karasuk_RISE496 3,57491
AAC_Karasuk_outlier_RISE495 3,45846
AAB_Karasuk_RISE499 3,23068
AAB_Anglo_Saxon_NO3423 3,151901
AAB_Portugal_MN_LugarCanto44 2,841017
AAB_Ireland_MN_Ballynahatty_BA64 2,83384
AAB_Hungary_HG_I1507 2,83234
AAC_Altai_IA_RISE602 2,299445
AAB_Spain_LNCA_ATP16 2,297892
AAB_Portugal_LNCA_CovaMoura364 2,20875
AAB_Sweden_MN_Gok2 2,18115
AAB_Satsurblia_SATP 2,1234
AAB_Anatolia_N_Bar31 2,12068
AAB_Roman_Britain_6DRIF-21 2,11004
AAB_LaBrana1_LaBrana1 1,97584
AAB_Portugal_MBA_MonteGato104 1,88084
AAB_Loschbour_Loschbour 1,65013
AAB_Unetice_RISE150 1,57885
AAB_Greece_LN_Klei10 1,56901
AAB_Portugal_MN_LugarCanto42 1,49954
AAB_Spain_LNCA_Matojo 1,402323
AAB_Kotias_KK1 1,327316
AAB_Portugal_MBA_TV3831 1,32433
AAB_Anatolia_N_Bar8 1,12844
AAB_LBK_Stuttgart 1,0844441
AAB_Hungary_BA_RISE479 0,881237
AAC_Altai_IA_RISE504 0,808801
AAB_Bichon_Bichon 0,5946
AAB_Greece_EN_Rev5 0
AAB_Greece_LN_Pal7 0
AAB_Hungary_BA_I1504 0
AAB_Hungary_CA_I1497 0
AAB_Hungary_IA_IR1 0
AAB_Hungary_N_I1495 0
AAB_Hungary_N_I1496 0
AAB_Motala_HG_Motala12 0
AAB_Oetzi_Oetzi 0
AAB_Portugal_LNCA_DolmenAnsiao96B 0
AAB_Roman_Britain_outlier_3DRIF-26 0
AAC_Altai_IA_RISE601 0
AAC_Karasuk_outlier_RISE493 0
AAC_Karasuk_outlier_RISE497 0

Tesmos said...

Top 50 Modern matches:

English_Kent_HG00141 27,7478522
Utah_German_NA12239 24,446229
English_Kent_HG00138 19,730626
Estonian_ee59 17,001032
English_Kent_HG00152 14,469127
Orcadian_HG00110 14,255506
French_HGDP00514 14,071683
Utah_German_NA12057 13,181542
English_Kent_HG00145 12,76463
English_Cornwall_HG00243 12,7308841
Orcadian_HG00121 12,458477
English_Kent_HG00151 12,333112
Ashkenazi_Jew_ashkenazy7e 12,188303
Bulgarian_Bulgarian10H 12,1013928
Orcadian_HGDP00800 12,076324
English_Kent_HG00135 11,845454
Montenegrin_Montenegro21_GSM1424649 11,75365949
Utah_USA_NA11881 11,709575
English_Kent_HG00139 11,3820507
English_Cornwall_HG00263 11,37624
Spanish_Pais_Vasco_HG01521 11,282587
Utah_German_NA07000 10,976353
English_Kent_HG00146 10,969149
French_HGDP00538 10,92928
Croatian_GSM1841126 10,85597
Norwegian_NOR111 10,724372
Estonian_ee105 10,65209731
Utah_USA_NA12234 10,448183
Estonian_GS000017209 10,292224
English_Kent_HG00142 10,28596
English_Cornwall_HG00251 10,1338139
Utah_German_NA12716 10,1221904
Utah_USA_NA12749 10,094053
Utah_USA_NA11995 10,086597
French_HGDP00537 10,019936
Utah_USA_NA11892 9,9801962
English_Kent_HG00128 9,908737
Utah_USA_NA12056 9,875564
Belarusian_belorus5 9,833302
Lithuanian_lithuania5 9,735358
Russian_North_HGDP00889 9,636861
Orcadian_HGDP00810 9,6295
Ukrainian_West_UkrLv228 9,569684
Norwegian_NOR108 9,5295761
Utah_British_NA12827 9,289007
Orcadian_HG00123 9,287136
Cossack_Ukrainian_GS000035238 9,263278
Belarusian_belorus7 9,248209
Norwegian_NOR106 9,242999

huijbregts said...

My ancients are easy to interpret. Among the top 11 I have three samples from Roman Britain Driffield Terrace.
As I am South_Dutch/Belgian, it seems plausible that they had a lot of Belgae ancestry.

AAB_Roman_Britain_6DRIF-3 21,333273
AAB_Yamnaya_Kalmykia_RISE548 14,08972
AAB_Roman_Britain_6DRIF-21 13,231303
AAB_Unetice_RISE577 13,140402
AAB_Ireland_EBA_Rathlin2_RSK1 11,2042198
AAB_Sintashta_RISE395 10,910651
AAB_Anglo_Saxon_NO3423 10,792951
AAB_Nordic_IA_RISE174 9,276697
AAB_Portugal_MN_LugarCanto42 8,6604467
AAB_Ireland_MN_Ballynahatty_BA64 8,619298
AAB_Roman_Britain_6DRIF-22 8,25345
AAB_Karasuk_RISE496 6,6589617
AAB_Afanasievo_RISE511 5,93304
AAB_Andronovo_RISE505 5,835642
AAB_Hungary_BA_I1504 5,50697
AAB_Yamnaya_Kalmykia_RISE552 4,90714
AAB_Unetice_RISE150 4,7994111
AAB_Portugal_LNCA_DolmenAnsiao96B 4,765008
AAB_Portugal_MBA_MonteGato104 4,4965848
AAB_Portugal_MBA_TV3831 4,3699011
AAC_Karasuk_outlier_RISE495 4,2248325
AAB_Nordic_LN_RISE98 4,205276
AAB_Motala_HG_Motala12 4,084636
AAB_Portugal_MN_LugarCanto44 4,053803
AAB_Loschbour_Loschbour 4,0308263
AAB_Iberia_EN_CB13 3,995065
AAB_LaBrana1_LaBrana1 3,920796
AAB_Portugal_LNCA_CabecoArruda122A 3,9081
AAB_Spain_LNCA_ATP16 3,61109
AAB_Andronovo_RISE500 3,59557
AAB_Oetzi_Oetzi 3,46608
AAB_Hungary_CA_I1497 3,350283
AAC_Altai_IA_RISE504 3,268279
AAB_Hungary_N_I1506 3,087775
AAB_Bichon_Bichon 3,0437467
AAB_Karasuk_RISE499 2,754702
AAB_Roman_Britain_6DRIF-18 2,753474
AAB_Ireland_EBA_Rathlin1_RM127 2,665477
AAC_Karasuk_outlier_RISE493 2,5583
AAB_Slav_Bohemia_RISE569 2,381255
AAB_Hungary_BA_RISE479 2,299432
AAB_Portugal_MN_LugarCanto41 2,229835
AAB_LBK_Stuttgart 2,156556
AAB_Hungary_HG_I1507 2,07099
AAC_Altai_IA_RISE602 2,054522
AAB_Pitted_Ware_Ajv58 1,9885
AAB_Mezhovskaya_RISE523 1,96248
AAB_Spain_LNCA_Matojo 1,86028
AAB_Kotias_KK1 1,6344
AAC_Karasuk_outlier_RISE502 1,5817364
AAB_Portugal_MBA_TV32032 1,3784922
AAB_Sweden_MN_Gok2 1,137183
AAA_Ger_Huijbregts_Ger_Huijbregts 1
AAB_Greece_LN_Pal7 0,968163
AAB_Anatolia_N_Bar31 0
AAB_Anatolia_N_Bar8 0
AAB_Greece_EN_Rev5 0
AAB_Greece_LN_Klei10 0
AAB_Hungary_IA_IR1 0
AAB_Hungary_N_I1495 0
AAB_Hungary_N_I1496 0
AAB_Portugal_LNCA_CovaMoura364 0
AAB_Portugal_LNCA_CovaMoura9B 0
AAB_Roman_Britain_outlier_3DRIF-26 0
AAB_Satsurblia_SATP 0
AAB_Spain_LNCA_ATP2 0
AAC_Altai_IA_RISE601 0
AAC_Karasuk_outlier_RISE497 0

jv said...

What DNA test do I need to take to have to have this done?( I've only had the full sequence MtDNA test) I'm interested in the comparison with ancient populations. When do you plan to do this again? Thanks! (......even though my MtDNA is post-LBK, I have a lot of German ancestry from Venne and those women were GAC & LBK)

Davidski said...

@jv

You need an autsomal data file from FTDNA, Ancestry or 23andMe. They usually go for around $100 these days.

jv said...

Thank you. I will order from FTDNA as I have my MtDNA account with them. COULD you please give me your opinion regarding my MtDNA lineage? ANY idea when MtDNA H6a1 entered the PC Steppe? Do you think H6a1 was in the pre-Yamnaya populations? Also, did she come from Mesolithic Iran/Caucasus? Maybe the Zarzian Culture? I can trace the line H6 from Israel 4000 BCE to H6a1a Esperstedt , Germany 2300 BCE Schnurkeramikkultur,1800 BCE Netherlands Glockenbecherkultur and H6a1a 1800 BCE Srubna Culture Samara, Russia. Thank you

Dmytro said...

"I simply loaded my "ancients_only" file into Past 3, selected all of the columns and rows, and then did this: Multivariate > Clustering > Neighbor joining. Note that I cluster on the same branch as Slav_Bohemia, and this makes perfect sense considering my Polish ancestry. By the way, I dropped Oetzi from this run because he was behaving strangely, which is not unusual for low coverage genomes. Click on the image and open in a new tab for a better view."

This didn't seem to work for me. When I clicked after "Neighbour joining" I was told that the first two rows (in effect since I had to experiment with removals. Anything less than these two rows as a whole was constantly termed erroneous) were errors. When I removed them I did get a result, but the clusters included only the ancient samples and not my figures.

Derek said...

I'm a NW European, but I also clustered with Slav Bohemia on the neighbor joining tree.

Davidski said...

I'm a NW European, but I also clustered with Slav Bohemia on the neighbor joining tree.

This does happen occasionally, probably because Slav_Bohemia is the most recent ancient genome in the analysis. Different clustering methods might show different results as well.

Tesmos said...

@Davidski,

Should I use an ancient or a relevant average modern population for the Generalized Linear Model?

Davidski said...

Why not try both?

Matt said...

With the 3000+ samples list, I haven't been able to do much, but thought the following PCA exercise may be of interest to some:

Part1:

a) Separated all the samples by population grouping*: https://pastebin.com/v4d4Bm5b

b) Take a list of populations which have reasonably high levels of total sharing with the ancients (including some West Eurasians as well that don't): https://pastebin.com/Cvyx2Udu

(My cutoff is pretty arbitrary here)

c) Then after getting that list, using just the ancients as columns and modern people as rows, produce a PCA:

Result is this: https://ibb.co/jPTHJk

Each population's label sits at the centre of where its dots (individual samples) plot - and there's a lot of variation within each population here.

First PC1 separates populations which are generally high tract length sharers with the ancients from populations which are low. This dimension spans the British Isles+Norway on the positive end, as the highest sharers cumulatively with all ancients, down to Near East on the negative with very little sharing. Most of the rest of Europe about 50-75% the distance from the Near East to BI on this dimension.

(A point of interest to me is that the British Isles is distinct here from populations who generally look to have similar proportions of WHG:Anatolian:Steppe, like Slovakian or German, and overlap heavily with one another. This does not mean that they have more ancient European ancestry or something but is due to the ancients including BI ancients and generally from the whole region from Iberia->Germany, and more likely to be specifically ancestral.).

Second PC2 then splits populations with relatively high Iberia farmer sharing and lower Steppe and Karasuk (e.g. most extreme Sardinian) sharing on the other from populations who are the reverse (e.g. most extreme Kets).

* For an alternative list of population groupings: https://pastebin.com/TCbjTbiJ. I've merged a lot of the populations here, to make it easier for me to run averages.

Matt said...

Part 2: So, variations on that PCA above:

A) Removing row populations who have high sharing with Karasuk, and pretty low sharing with Iberia_EN (e.g. Kets etc.) - https://ibb.co/cnmHZQ, https://ibb.co/fhrvM5, https://ibb.co/hJnPEQ

That gives a more decompressed look at the variation in the European populations without hugely changing the PCs.

(Population list: https://pastebin.com/D2pwKHBB)

B) Modified A) by removing column ancients from the Iron Age (Roman Britain, Anglo Saxon, Slav Bohemia) - https://ibb.co/gTYEW5, https://ibb.co/cctEW5

Removed in case the preponderance of Iron Age to CE samples from Britain were swelling up the BI position in PC1. Does appear to have reduced that effect, however the graph is mostly the same, so this shape doesn't only or mainly come from that.

C) Modified B) by removing column ancients from the Irish EBA as well : https://ibb.co/n4geyk, https://ibb.co/jkbCJk

Flattens things out once again, however again, no major changes.

Going back to A) though, if we do include the Iron Age populations and go to PC3: https://ibb.co/jR9or5, and loadings there: https://ibb.co/enQiPQ
This PC3 splits a tendency on the posistive end to share CM with the most recent ancestors of Eastern European - HungaryBA and Slav Bohemia - from a more general set of affinities to the ancients in general on the negative end of the PC. Once decent quality samples from the largely Steppe_EMBA+Narva Baltic_BA population are available, I expect they would also load heavily on the positive end of a similar PC.

Contrasting the position of the Tajik populations (and apparently Uralic speaking Udmurds) are also pretty interesting here, fitting relatively close to Eastern Europe in the overall structure of PC1 and PC2, due to both having relatively high steppe related CM sharing and lower CM sharing with ancient Iberians and North-Central Bronze Age Europeans... and then very distinct in PC3 because, of course, not sharing these more recent ancestors from late Bronze Age to late Iron Age / CE with present day Eastern European people.

Matt said...

Part 2: So, variations on that PCA above:

A) Removing row populations who have high sharing with Karasuk, and pretty low sharing with Iberia_EN (e.g. Kets etc.) - https://ibb.co/cnmHZQ, https://ibb.co/fhrvM5, https://ibb.co/hJnPEQ

That gives a more decompressed look at the variation in the European populations without hugely changing the PCs.
(Population list: https://pastebin.com/D2pwKHBB)

B) Modified A) by removing column ancients from the Iron Age (Roman Britain, Anglo Saxon, Slav Bohemia) - https://ibb.co/gTYEW5, https://ibb.co/cctEW5

Removed in case the preponderance of Iron Age to CE British samples were swelling up the BI position in PC1. Does appear to have reduced that effect, however the graph is mostly the same, so this shape doesn't only or mainly come from that.

Matt said...

Part 3:

C) Modified B) by removing column ancients from the Irish EBA as well : https://ibb.co/n4geyk, https://ibb.co/jkbCJk

Flattens things out once again, however again, no major changes.

Going back to A) though, if we do include the Iron Age populations and go to PC3: https://ibb.co/jR9or5, and loadings there: https://ibb.co/enQiPQ

This PC3 splits a tendency on the posistive end to share CM with the most recent ancestors of Eastern European - HungaryBA and Slav Bohemia - from a more general set of affinities to the ancients in general on the negative end of the PC.

Once decent quality samples from the largely Steppe_EMBA+Narva Baltic_BA population are available, I expect they would also be part of loading heavily on the positive end of a similar PC.

Contrasting the position of the Tajik populations and apparently Uralic Udmurds are also pretty interesting here, fitting relatively close to Eastern Europe in the overall structure of PC1 and PC2, due to having relatively high steppe related CM sharing and lower CM sharing with ancient Iberians and North-Central Bronze Age Europeans... and then very distinct in PC3 because, of course, due to not sharing these more recent ancestors from late Bronze Age to late Iron Age / CE with present day Eastern European people.

(At Davidski, sorry, an earlier version of this got trapped in your anti-link spam filter, so please delete that if you see).

Matt said...

One last thing also, same as above, using just Bronze Age or later ancients as columns:

PC1 / PC2: https://ibb.co/mAFiPQ (PC1: high Bronze Age or later sharing vs lower BA or later sharing, PC2: East vs West Bronze Age)

PC1 / PC3: https://ibb.co/mAFiPQ (PC3: Bronze Age or later ancestors or recent Eastern Europe vs other Bronze Age or later)

PC2 / PC3: https://ibb.co/jbnf4Q

Loadings: https://ibb.co/hYHf4Q

(Fewer columns so slightly more noise in position of populations with low n).

Matt said...

Had some time to use run population averages, just with the ancient individuals as columns for now:

1. Grouped Ancients as rows, individual ancients as columns: https://pastebin.com/YAVpGj0P

2. Grouped moderns as rows, individual ancients as columns: https://pastebin.com/YAVpGj0P

(Not sure if this is worthwhile or if you all have already run / got population averages.)

Anyway, some graphics based on 2. above: https://imgur.com/a/7F24w
All pretty self explanatory

(Btw, I think the method of running an average I have here may be slightly better than the one I used before, which I'm as sure about now. Apologies, folks who used that data).

Davidski said...

@jv

I don't know much about H6a1a.

But I'd say that you should be able to track the entry of your lineage into Central Europe with precision within a few years as many more ancient mito genomes are sequenced.

KVN said...

These are my ordered non-zero ancients

AAB_Sintashta_RISE395 13.468412
AAB_Unetice_RISE577 12.097242
AAB_Yamnaya_Kalmykia_RISE552 9.746889
AAB_Ireland_EBA_Rathlin1_RM127 8.557313
AAB_Unetice_RISE150 8.12866
AAB_Hungary_BA_RISE479 8.10181
AAC_Karasuk_outlier_RISE502 7.581374
AAB_Karasuk_RISE499 7.27049671
AAB_Yamnaya_Kalmykia_RISE548 6.908238
AAB_Greece_EN_Rev5 6.541456
AAB_Ireland_MN_Ballynahatty_BA64 6.416608
AAB_Afanasievo_RISE511 5.768278
AAB_Sweden_MN_Gok2 5.752744
AAB_Anglo_Saxon_NO3423 5.715692
AAB_Loschbour_Loschbour 5.132116
AAB_Andronovo_RISE505 4.88297887
AAB_Roman_Britain_6DRIF-3 4.785061
AAB_LBK_Stuttgart 4.7076274
AAB_Hungary_N_I1496 4.68666
AAB_Portugal_LNCA_CovaMoura364 4.29161105
AAB_Bichon_Bichon 4.206344
AAB_Ireland_EBA_Rathlin2_RSK1 4.093057
AAB_Portugal_MN_LugarCanto42 3.977782
AAB_Pitted_Ware_Ajv58 3.85059
AAB_Nordic_LN_RISE98 3.594694
AAB_Roman_Britain_6DRIF-22 3.346546
AAB_Roman_Britain_6DRIF-18 3.0715997
AAC_Altai_IA_RISE601 2.831298
AAB_Anatolia_N_Bar31 2.707102
AAB_Karasuk_RISE496 2.69625
AAB_Mezhovskaya_RISE523 2.6578239
AAB_Portugal_MBA_MonteGato104 2.589931
AAB_Portugal_MN_LugarCanto41 2.429286
AAB_Greece_LN_Klei10 2.4218
AAB_Slav_Bohemia_RISE569 2.31379
AAB_Portugal_MN_LugarCanto44 2.30726
AAB_Greece_LN_Pal7 2.176416
AAB_Roman_Britain_outlier_3DRIF-26 2.084594
AAB_Portugal_LNCA_CabecoArruda122A 2.03082
AAB_Satsurblia_SATP 1.863605
AAC_Karasuk_outlier_RISE493 1.791589
AAB_Hungary_IA_IR1 1.78818
AAB_Roman_Britain_6DRIF-21 1.702548
AAB_Oetzi_Oetzi 1.675735
AAB_Spain_LNCA_ATP2 1.590423
AAC_Altai_IA_RISE602 1.56451
AAB_Iberia_EN_CB13 1.451491
AAB_Andronovo_RISE500 1.439435
AAB_Hungary_CA_I1497 1.390123
AAB_Hungary_N_I1495 1.374484
AAB_Portugal_LNCA_DolmenAnsiao96B 1.32424
AAB_Anatolia_N_Bar8 1.305
AAB_Kotias_KK1 1.264201
AAC_Karasuk_outlier_RISE495 1.24146
AAB_Hungary_BA_I1504 1.120406
AAB_Portugal_MBA_TV3831 0.893657
AAB_Portugal_LNCA_CovaMoura9B 0.726186

Samuel Andrews said...

A lot of the results posted here make sense. Iberia Chalcolithic and modern Spain matching with a Spaniard, Nordin_LN with Norwegian, Roman Britain for England and NW France. But if one's ancestry is so mixed I doubt this test is of much use. Interestingly, Yamnaya seems to keep popping up for NW Euros. We'd have to see other European results to see if this is a legitimate correlation.

@David,

Could something like this resolve genealogical relationships between Olalde and Mathiesons's ancient genomes? Like who was North Bell Beaker's Neolithic ancestor?

@jv,

This is the age estimate I get for H6a1.

H 15ky
---H6 12ky
------H6a 10ky
---------H6a1 10ky
------------H6a1a 6ky
------------H6a1b 8ky

"Steppe" could have formed between 5000 and 4000 BC which several thousand years after these age estimates. I'm thinking H6a1 had already formed and expanded in the Caucasus before coming onto the Steppe.

Helgenes50 said...

@ Samuel Andrews

Yes, my results as NW French and Norman make sense
Old Britons, Nordic and Anglo-Saxons are at the top of my list.

AAB_Sintashta_RISE395 16.7517755
AAB_Roman_Britain_6DRIF-22 12.372527
AAB_Unetice_RISE577 11.72396279
AAB_Unetice_RISE150 11.3984053
AAB_Anglo_Saxon_NO3423 10.795745
AAB_Portugal_LNCA_DolmenAnsiao96B 10.53239
AAB_Afanasievo_RISE511 9.507238
AAB_Nordic_IA_RISE174 8.849702
AAB_Mezhovskaya_RISE523 8.82688
AAC_Karasuk_outlier_RISE502 8.078086
AAB_Yamnaya_Kalmykia_RISE552 6.857887

Chad Rohlfsen said...

What Norwegian?

Matt said...

So had a go at calculating population averages for length CM sharing.

It does seem that they vary hugely between population, going from 3806(!) in Nganasan down to 2 in Cypriot. Values are huge in modern Siberians, even by comparison to Bronze Age and HG ancients. Some of these groups are compromised of relatively close relatives, particularly the small Siberian groups - surprised me that these were way high than Karitiana who look very drifted in Fst scores, and isolated maritime Oceanian populations.

Europeans generally have very short or average sharing averages within populations (which makes sense for huge and admixing populations) in the range of 2 - 20. There are some clear outliers in Russian Central (288) and Cossack Ukrainian (162) who may have been sampled from relatively close relatives (not close enough like first order relatives to distort unlinked PCA, but close enough to have a high excess of sharing here).

The sheer diversity of value sizes looks to make it impossible just to run a PCA on these. Even in neighbour joining, the branch length differences are huge. Using lengths is super sensitive to very recent shared relatedness, which can be significant in panels where they're only screening out very close relatives (because it's not important to the work they're doing), or it's not possible to do that because the populations are so small. (This is why chunkcounts can be more useful for PCA in some circumstances than the length information?).

Anyway, here's 1) the .CSV of all pop averages so anyone's free to try with PAST3 though: https://file.io/mNsALZ

Also 2), here's a .CSV modern groups for rows and grouped ancient averages for columns:

https://file.io/0dK6fy

Some more graphs with the ancients as columns and moderns as rows, using group averages in rows and columns, which smooths out a lot that is much noisier with row and column individuals:

https://imgur.com/a/8mwv6
https://imgur.com/a/YyckG

Open Genomes said...

How would this work compared to running 100 SNPs and 3 cM against these samples in Gedmatch?

Also, with phased SNP array data, many of the smaller segments tend to vanish. That's not "IBS vs. IBD", just "unphased data and faulty results".


There's another interesting thing you can do, too:
Use the DNA.land ~39 million imputed SNPs instead of the typical arrays.

We've uploaded to Bar8 from Barcin NW Anatolia c. 8300 BP (no relatives; educational attainment: 14.7 years You have attended college ;) ) and WC1 from Wezmeh Cave, Iran (no relatives either; 13.8 years You have attended college :D ) so I can download the VCF files and make them available, although I think already these have excellent coverage at the typical 1.2 million combined SNP array chip SNPs. (That includes the Affymetrix Human Origins Array as well as the Illumina chips.)


Do we really know the IBD segments at the whole genome sequence level that constitute "WHG, Anatolian Neolithic, Iranian Neolithic" etc.?

What do we know about IBD between some of these high-coverage ancient whole genomes and modern whole genome sequences, for example from the 1K Genomes and the SGDP? Does this look the same as IBD using the "minimal set' of SNPs? What about very rare alleles?

Also how about using the Southern and Eastern African aDNA samples, and also waiting for the aDNA North Africans?

Davidski said...

This is probably more of a haplotype test than an IBD test, but whatever it is, it's more informative for fine scale ancestry than using unlinked markers (thus, much better than IBS, for example), and indeed very accurate. The accuracy of the output can be seen in these PCA.

http://eurogenes.blogspot.com.au/2014/08/haplotype-based-pca-of-west-eurasia-and.html

Also, unlike IBD tests elsewhere, there is a lot of already integrated output, so users don't have to ask other users for their results, or try to run reference samples themselves, to put their own results into a context.

However, I really can't just add ancient samples from different sources. It's best if all of the ancient samples are genotyped and prepped in exactly the same way.

Tesmos said...

I ran 2 Generalized Linear Models with the Anglo-Saxon and the Nordic IA RISE174. It looks like I share somewhat more ancestry with EBA Steppe peoples.

Anglo-Saxon

https://drive.google.com/file/d/0B-tlHcv5HwZBdEVnTTJha1p5N0k/view?usp=sharing

Nordic IA

https://drive.google.com/file/d/0B-tlHcv5HwZBTWhmNUZWaVJjVVk/view?usp=sharing

Davidski said...

Please note that I'll be doing another run at the end of next month. It'll be exactly the same as the last one, so no point doing it again for anyone who's already done it.

Interestingly, what I found is that this test is very good at picking up not only minor Ashkenazi ancestry, but also minor Sephardic Jewish ancestry, so if anyone is interested in exploring this possibility then taking part in this exercise should be very useful.

Matt said...

Possibly of interest (not sure if anyone read the other posts): I was having a look again at the matrix of group averages.

Using the simple matrix of averages produces PCA and neighbour joining which don't show very much. Or rather they do show structure, but because the Nganasan and ancient related dimensions are so high, other structure is very small in the PCA / Neighbour Joining:

So: https://imgur.com/a/NcOwS

Thinking about this, I carried out a square root transformation of the matrix. That would make the covariance between large (e.g. at most extreme Nganasan) and small dimensions (at most extreme, Cypriot) a lot more even.

Running PCA and Neighbour Joining on the square root of the average shared CM matrix:

https://imgur.com/a/qevWQ

This gives something which is more reflective of deep structure and less dominated by the recent structure.

This process, using the square root of the matrix / CM may be useful if you want to look at a slightly different balance of the recent and ancient structure (downside is that this isn't really interpretable in the same straightforward way).

(Various graphs using the square root of shared CM: https://imgur.com/a/XhLG3).

Gabriel Opi said...

Here are my anicents:

AAB_Afanasievo_RISE511 2.09049

AAB_Anatolia_N_Bar31 0

AAB_Anatolia_N_Bar8 0.97035

AAB_Andronovo_RISE500 4.512618

AAB_Andronovo_RISE505 6.612144

AAB_Anglo_Saxon_NO3423 7.49012

AAB_Bichon_Bichon 0

AAB_Greece_EN_Rev5 0

AAB_Greece_LN_Klei10 1.683009

AAB_Greece_LN_Pal7 3.26847974

AAB_Hungary_BA_I1504 6.296164

AAB_Hungary_BA_RISE479 1.82977

AAB_Hungary_CA_I1497 1.023273

AAB_Hungary_HG_I1507 3.463794

AAB_Hungary_IA_IR1 0

AAB_Hungary_N_I1495 2.465896

AAB_Hungary_N_I1496 0

AAB_Hungary_N_I1506 3.62879

AAB_Iberia_EN_CB13 3.682671

AAB_Ireland_EBA_Rathlin1_RM127 8.88122

AAB_Ireland_EBA_Rathlin2_RSK1 7.1050328

AAB_Ireland_MN_Ballynahatty_BA64 3.161594

AAB_Karasuk_RISE496 4.10343

AAB_Karasuk_RISE499 3.2259

AAB_Kotias_KK1 0

AAB_LaBrana1_LaBrana1 0

AAB_LBK_Stuttgart 2.9941529

AAB_Loschbour_Loschbour 2.933235

AAB_Mezhovskaya_RISE523 11.743162

AAB_Motala_HG_Motala12 0

AAB_Nordic_IA_RISE174 3.050938

AAB_Nordic_LN_RISE98 13.4329

AAB_Oetzi_Oetzi 0

AAB_Pitted_Ware_Ajv58 0

AAB_Portugal_LNCA_CabecoArruda122A 0.87258

AAB_Portugal_LNCA_CovaMoura364 1.21213

AAB_Portugal_LNCA_CovaMoura9B 2.92982

AAB_Portugal_LNCA_DolmenAnsiao96B 1.796368

AAB_Portugal_MBA_MonteGato104 1.615219

AAB_Portugal_MBA_TV32032 6.447425

AAB_Portugal_MBA_TV3831 0

AAB_Portugal_MN_LugarCanto41 0

AAB_Portugal_MN_LugarCanto42 3.16183

AAB_Portugal_MN_LugarCanto44 4.03809

AAB_Roman_Britain_6DRIF-18 7.89825

AAB_Roman_Britain_6DRIF-21 4.011221

AAB_Roman_Britain_6DRIF-22 5.06692

AAB_Roman_Britain_6DRIF-3 9.356601

AAB_Roman_Britain_outlier_3DRIF-26 0

AAB_Satsurblia_SATP 0

AAB_Sintashta_RISE395 6.449945

AAB_Slav_Bohemia_RISE569 4.428073

AAB_Spain_LNCA_ATP16 2.3384

AAB_Spain_LNCA_ATP2 2.183731

AAB_Spain_LNCA_Matojo 2.818434

AAB_Sweden_MN_Gok2 5.690385

AAB_Unetice_RISE150 4.736982

AAB_Unetice_RISE577 7.486465

AAB_Yamnaya_Kalmykia_RISE548 12.847587

AAB_Yamnaya_Kalmykia_RISE552 8.945287

AAC_Altai_IA_RISE504 0

AAC_Altai_IA_RISE601 2.60782

AAC_Altai_IA_RISE602 0

AAC_Karasuk_outlier_RISE493 0

AAC_Karasuk_outlier_RISE495 5.870795

AAC_Karasuk_outlier_RISE497 0

AAC_Karasuk_outlier_RISE502 6.4392894

Gabriel Opi said...

Top to 5.
Modern Matches
(only included top of English and left out Utah)

Estonian_ee14 20.519316
Estonian_ee147 17.7388588
Hungarian_GS000016901 15.0727931
English_Kent_HG00131 12.5047673
English_Cornwall_HG00254 12.2043229
Norwegian_NOR111 11.9545958
English_Cornwall_HG00233 11.673633
Orcadian_HG00117 11.3710984
Orcadian_HG00110 10.9251329
Estonian_GS000017199 10.246228
Russian_North_HGDP00881 9.658181
Finnish_GS000016894 9.645945
Macedonian_Macedonian19_GSM1424617 9.518464
Orcadian_HG00115 9.275291
English_Cornwall_HG00234 9.272401.3255389
Hungarian_hungary7 9.948052
Orcadian_HG00123 9.03231
Slovakian_Slovakia150 9.0809014
Chuvash_Chuvash11 8.726197
Slovakian_Slovakia429 8.668007
Russian_North_HGDP00894 8.6256213
Mari_GS000014331 8.596761
English_Cornwall_HG00252 8.299781
Ukrainian_East_UkrBel618 8.14323
Ukrainian_East_GS000035121 8.1435916
Romanian_Romania13 8.146691
Adygei_HGDP01384 7.9683
Russian_North_HGDP00896 7.932111
Belarusian_belorus7 7.605246
Erzya_mordovia11 7.485887
Scottish_Argyll_HG00103 7.27218
Ukrainian_East_GS00003512 8.1435916
Romanian_Romania13 8.146691
French_HGDP00530 7.873901
English_Cornwall_HG00261 7.905423
Spanish_Extremadura_HG01531 7.712461
Spanish_Spanish33H 7.705787
Spanish_Murcia_HG01756 7.758243
Erzya_mordovia11 7.485887
Ukrainian_North_GS000013755 7.377048
Spanish_Galicia_HG01704 6.829332
Polish_Polish16H 6.786185
English_Cornwall_HG00260 6.652127
Polish_Polish11H 6.584474
rench_HGDP00527 6.5892593
Serbian_Bosnia_Serbian_B-H9_GSM1424669 6.540832
Montenegrin_Montenegro1_GSM1424636 6.503154
Chuvash_Chuvash13 6.431465
Latvian_latvian58C8 6.412131
Swedish_GS000035240 6.408373
Spanish_Valencia_HG01607 6.356098
Greenlander_West_GSM558765 6.30466
Estonian_ee136 6.336066
Estonian_ee72 6.263758
French_HGDP00511 6.254699
Spanish_Spanish34H 6.083324
Puerto_Rican_GSM531197 6.0123951
Spanish Aragon_HG01676 6.389606
Adygei_HGDP01383 6.227883
Norwegian_NOR148 6.0034178
Bosnian_Bosnian_8_GSM1424664 5.9583
Belarusian_GS000035241 5.951467
Basque_French_HGDP01377 5.559201
Tajik_Ishkasim_GS000016189 5.85127345
English_Cornwall_HG00256 5.841081
Kyrgyz_GS000016191 5.820921
Spanish_Cantabria_HG01513 5.808557
French_HGDP00516 5.666612
Lithuanian_lithuania2 5.575585
Armenian_Martuni_594711 5.485525
Uzbek_uzbek88 5.406702
Greenlander_West_GSM558764 5.404897
BedouinA_HGDP00635 5.313244
Croatian_GSM1841134 5.152686375
Chuvash_Chuvash6 5.090122
Selkup_selkup2 5.0563243