Southern Italian Ethnogenesis (My theory)

Er Monnezza said:
Results look different because I used averages instead of aggregated individual samples. Now mine should be more in line with the other regions.

SampleMinoanYamnayaRemedelloAnatolian_BAC_Italian_NIberomaurusianC_Italian_ChLBolshoy_Ostrov
Italian_Friuli_VG33,934,431,100000,6
Italian_Trentino28,733,737,500000,1
Italian_Aosta_Valley20,933,744,6000,600,2
Italian_Veneto36,331,931,700000,1
Swiss_Italian28,731,439,900000
Italian_Piedmont33,130,135,7000,900,2
Italian_Lombardy35,428,236,300000,1
Italian_Lazio4628,12,7022,40,800
Italian_Emilia37,627,726,606,70,210,2
Italian_Tuscany40,327,422,408,80,110
Italian_Romagna42,927,214,5013,30,31,80
Italian_Umbria47,426,812,3011,10,81,40,2
Italian_Liguria39,126,732,5001,600,1
Italian_Marche50,526,35,6016,10,80,70
Italian_Apulia64,124,60901,600,7
Italian_Abruzzo68,623,9051,90,500,1
Italian_Molise72,622,104,90,30,100
Italian_Basilicata57,522,1018,601,800
French_Corsica41,421,122,6013,9100
Italian_Sicily64,620011,20,33,900
Italian_Campania6919,809,80,31,100
Ajeje_Brazorf_imputed62,218,7012,92,33,10,60,2
Ajeje_Brazorf65,916,7013,104,300
Italian_Calabria51,114,4031,702,800
Italian_Jew47,79,403507,900
Sardinian331,153,508,83,600
Click to expand...

These are the Updated Dodecad samples, I had actually done another model that is more universal for Europeans. These samples all have a good fit with the model, notice how the south still prefers Minoan though. The simpler K8 model I made was more targeted towards modeling south Italians specifically. It should also be noted that the fit becomes very bad for others, such as modern Near Easterners:

ko72RYU.png
 
Salento said:
I’m in Italy :) … I don’t have full access to my files, … I’ll take a look when I get back.
Click to expand...

Salento: Congrats. (y) Enjoy your trip. How is the local vinno and pasta? I am sure it is great, when I was in Rome and Sicily in 2019, I did not have a bad meal and I was there for over 3 weeks.
 
Jovialis said:
These are the Updated Dodecad samples, I had actually done another model that is more universal for Europeans. These samples all have a good fit with the model, notice how the south still prefers Minoan though. The simpler K8 model I made was more targeted towards modeling south Italians specifically. It should also be noted that the fit becomes very bad for others, such as modern Near Easterners:

ko72RYU.png
Click to expand...
Where can I find the coordinates to your model to check my results?
 
Jovialis said:
I posted this graphic previously that shows the full set of samples produced by Salento and Pax from the academic samples. They're the ones who labeled it. The graphic below is organized by region, the other graphic in the OP shows all the samples that have at least some Anatolia_BA on a cline to the most. As for TSI, I'm not sure if they had produced the individual samples.

a6Fvhei.png
Click to expand...

The samples labeled Sardinian, which are the HGDP panel chosen by Cavalli Sforza are extremely diverse. Three of the seven have NO Yamnaya and NO Minoan, Anatolian Bronze Age or Ibero-Marusian. He clearly hit the area on the highlands which is most remote but then got a few from surrounding areas which were a little bit more impacted. Still, the Minoans are a small fraction overall and Yamnaya almost non-existent.

So, as I always maintained, on those uplands there are still people who are pretty darn close to the pre-Bronze Age migrations into Europe.

I don't know where ITS or some of the West Sicily samples come from. Even the samples from Behar for West Sicily are very diverse in terms of Anatolia Bronze Age. One might say, ok, perhaps each Sicilian city has a unique population history, but there's variation even within the cities, i.e. the Busby Siracusa and Trapani numbers.

The Central Sicily, East Sicily, and Calabria samples are a bit more uniform, but then they are much fewer in number, so I don't think I can draw a hard and fast conclusion from that.

The Basilicata numbers, from what source I don't know, are all over the place again. It doesn't make sense. It's not even a large province, although it's certainly very mountainous and isolated, the place where even Christ couldn't reach. One would think that drift might have affected uniparentals, but autosomal should, perhaps, be rather uniform.

Campania and Apulia are also very diverse. I can understand it for Apulia, given it's length, but Campania? Unless, perhaps, some samples came from the city and others from the interior.

That's why it's important to know the academic paper used and how the samples were taken.

Except for one sample, the Tuscan HGDP are pretty uniform, and now I do see one sample for TSI, which is probably an average. It would be interesting to see the results for all 30 at some point. There's also a series of Tuscans for whom I don't know the provenance.

The northern samples are pretty uniform.

So, it's the southern samples where we see unexpected and unexplained diversity. We should ask Salento about it when he returns and see where some of these come from.

Meanwhile, I'm going to search through Busby and Behar to see how and where specifically their samples were selected.
 
Er Monnezza said:
Results look different because I used averages instead of aggregated individual samples. Now mine should be more in line with the other regions.

SampleMinoanYamnayaRemedelloAnatolian_BAC_Italian_NIberomaurusianC_Italian_ChLBolshoy_Ostrov
Italian_Friuli_VG33,934,431,100000,6
Italian_Trentino28,733,737,500000,1
Italian_Aosta_Valley20,933,744,6000,600,2
Italian_Veneto36,331,931,700000,1
Swiss_Italian28,731,439,900000
Italian_Piedmont33,130,135,7000,900,2
Italian_Lombardy35,428,236,300000,1
Italian_Lazio4628,12,7022,40,800
Italian_Emilia37,627,726,606,70,210,2
Italian_Tuscany40,327,422,408,80,110
Italian_Romagna42,927,214,5013,30,31,80
Italian_Umbria47,426,812,3011,10,81,40,2
Italian_Liguria39,126,732,5001,600,1
Italian_Marche50,526,35,6016,10,80,70
Italian_Apulia64,124,60901,600,7
Italian_Abruzzo68,623,9051,90,500,1
Italian_Molise72,622,104,90,30,100
Italian_Basilicata57,522,1018,601,800
French_Corsica41,421,122,6013,9100
Italian_Sicily64,620011,20,33,900
Italian_Campania6919,809,80,31,100
Ajeje_Brazorf_imputed62,218,7012,92,33,10,60,2
Ajeje_Brazorf65,916,7013,104,300
Italian_Calabria51,114,4031,702,800
Italian_Jew47,79,403507,900
Sardinian331,153,508,83,600
Click to expand...


thank you

mine below

Target: Torziok12b
Distance: 4.2477% / 4.24767897
36.7 Yamnaya
32.1 Minoan_Lasithi
31.2 Remedello
 
my father

Target: Ponsan_K12b
Distance: 6.4808% / 6.48080089
36.9 Yamnaya
33.2 Remedello
29.9 Minoan_Lasithi


seems like we come via Belluno province , Veneto .............between Friuli and Trentino
 
Salento said:
I’m in Italy :) … I don’t have full access to my files, … I’ll take a look when I get back.
Click to expand...

Stia attento a non sciogliersi col caldo di questi giorni...enjoy your trip nel Bel Paese
 
Angela said:
The samples labeled Sardinian, which are the HGDP panel chosen by Cavalli Sforza are extremely diverse. Three of the seven have NO Yamnaya and NO Minoan, Anatolian Bronze Age or Ibero-Marusian. He clearly hit the area on the highlands which is most remote but then got a few from surrounding areas which were a little bit more impacted. Still, the Minoans are a small fraction overall and Yamnaya almost non-existent.
Click to expand...

Sardinian HGDP are from Ogliastra.

Angela said:
Except for one sample, the Tuscan HGDP are pretty uniform, and now I do see one sample for TSI, which is probably an average. It would be interesting to see the results for all 30 at some point. There's also a series of Tuscans for whom I don't know the provenance.
Click to expand...

I do not think we know which 30 TSI individuals (Diekenes writes 21) were chosen by Metspalu to create the subset TSI30. The set originally consisted of over 100 samples (it is the one with at least 3 out of 4 grandparents born in Tuscany). A minority does not appear to be completely Tuscan, both by cross-referencing PCAs and their results. The average of all of them may be considered in line with other Tuscan samples. The other series of Tuscans might be another way of labelling the samples of Murlo, Casentino and Volterra. In some cases it is specified that they are from Murlo or Volterra. But originally there are many more than those released.


The results change somewhat if averages are used in the model instead of aggregated individuals.



TSI average based on about 98 TSI individuals (model in source with aggregated individuals by Jovialis). 48 out of 98 individuals get Anatolian_BA but they all get higher Central Italy Neolitich and lower Minoan. Cycles 4, Add Col Dist 0.25

kB4XNlJ.png




TSI average based on about 98 TSI individuals (model in source with averages posted by Er Monnezza). Only 7 out of 98 individuals get Anatolian_BA. Cycles 4, Add Col Dist 0.25

S566GJY.png
 
Last edited:
Dushman said:
Where can I find the coordinates to your model to check my results?
Click to expand...

Code: 
Bolshoy_Ostrov:3473±87:BOO006:Lamnidis_2018,4.07,44.75,0,0,0,36.53,2.94,0,0,11.3,0,0.41
Yamnaya:RISE240:Allentoft_2015,26.6,1.31,0,0,7.93,64.16,0,0,0,0,0,0
Yamnaya:RISE546:Allentoft_2015,22.07,0,0,0,0,70.45,0,0,0,0,7.49,0
Yamnaya:RISE547:Allentoft_2015,26.54,0,0,0,10.09,63.38,0,0,0,0,0,0
Yamnaya:RISE548:Allentoft_2015,27.76,0,0.93,0,3.91,65.86,0,0,0,0,1.35,0.19
Yamnaya:RISE550:Allentoft_2015,23.64,0.86,0,0,9.61,64.98,0.74,0,0,0,0,0.17
Yamnaya:RISE552:Allentoft_2015,30.94,1.6,0,0.01,2.06,61.11,0.39,0,0,0,3.1,0.78
Yamnaya:I0231_SVP3:Haak_2015,28.48,2.19,0,0,1.43,59.76,2.73,0,0,0,4.53,0.89
Yamnaya:I0357_SVP5:Haak_2015,31.45,0.86,0,0.77,4.05,54.56,2.22,0,0,0,6.09,0
Yamnaya:I0370_SVP10:Haak_2015,24.66,6.11,0,0,0,59.29,0,0,0,0,9.32,0.62
Yamnaya:I0429_SVP38:Haak_2015,26.2,3.3,0,0,2.84,64.62,2.03,0,0,0,0,1
Yamnaya:I0438_SVP50:Haak_2015,23.9,4.62,0,0,0,60.95,1.32,0.31,0,0,8.64,0.26
Yamnaya:I0441_SVP54:Haak_2015,33.48,1.54,0,0,0,56.53,8.46,0,0,0,0,0
Yamnaya:I0443_SVP57:Haak_2015,28.03,2.76,0,0,5.5,59.49,0,0,0,0,3.22,1
Yamnaya:I0444_SVP58:Haak_2015,30.94,0.57,0,0,3.61,59.43,0,0.11,0,0,2.51,2.83
Yamnaya:I2105:Mathieson_2018,25.39,2.85,0,0,1.94,58.76,1.17,0,0,0,8.18,1.72
Yamnaya:I3141:Mathieson_2018,24.66,4.21,0,0,2.47,64.51,1.21,0,0,0,2.93,0
Remedello:RISE487:Allentoft_2015,0,0,1.36,0,75.55,12.12,0,0,0.84,0,9.88,0.24
Remedello:RISE489:Allentoft_2015,0,0,0,0,72.08,12.24,0,0,5.69,0,9.99,0
C_Italian_ChL:R4:Antonio_2019,0,0,2.77,0.12,60.45,3.99,0,0,10.54,0.02,21.81,0.32
C_Italian_ChL:R5:Antonio_2019,0,0,2.76,0,63.45,3.27,0,0,10.15,0,20.2,0.17
C_Italian_ChL:R1014:Antonio_2019,0,0,2.67,0,64.87,0.54,0,0,0,0,31.92,0
C_Italian_N:R2:Antonio_2019,0,0,3.61,0,54.5,0,0,0.17,12.44,0,29.12,0.16
C_Italian_N:R3:Antonio_2019,0,0,4.22,0,52.75,0,0,0,11.28,0.2,31.18,0.38
C_Italian_N:R8:Antonio_2019,0,0,2.42,0,53.24,0,0,0,14.2,0.44,29.46,0.24
C_Italian_N:R9:Antonio_2019,0,0,2.58,0,50.8,0,0,0,13.79,0,32.49,0.34
C_Italian_N:R10:Antonio_2019,0,0,3.35,0,53.37,0,0,0,13.76,0,29.52,0
C_Italian_N:R16:Antonio_2019,0,0,3.41,0,57.73,1.14,0,0,9.12,0.21,28.3,0.09
C_Italian_N:R18:Antonio_2019,0,0,3.72,0,51.41,0,0,0.09,10.24,0,34.54,0
C_Italian_N:R19:Antonio_2019,0,0,4.33,0,55.74,0.85,0,0,9.83,0,29.26,0
Minoan:Lasithi:I0070:Lazaridis_2017,0,0,0.62,0.55,37.53,0,0,0,15.52,0,45.7,0.09
Minoan:Lasithi:I0071:Lazaridis_2017,1.16,0,2.96,0,37.9,0.23,0,0,13.01,0,44.59,0.15
Minoan:Lasithi:I0073:Lazaridis_2017,0,0.05,2.92,0.4,36.38,0,0,0,13.48,0,46.7,0.07
Minoan:Lasithi:I0074:Lazaridis_2017,0.58,0,4.33,0,39.44,0,0,0,12.45,0,43.19,0
Minoan:Lasithi:I9005:Lazaridis_2017,1.52,0,5.68,0,37.33,0,0,0,16.14,0,39.25,0.08
Minoan:Odigitria:I9130:Lazaridis_2017,1.38,0,0.19,0,41.77,0,0,0.28,17.61,0,38.77,0
Minoan:Odigitria:I9131:Lazaridis_2017,5.16,0,0,0,36.32,0,0,0,19.19,0,39.01,0.33
Minoan:Petras_EBA:Pta08:Clemente_2021,0,0,3.85,0.33,34.73,0.38,0.14,0.05,14.68,0.48,43.65,1.72
Anatolian_BA:I2495:Lazaridis_2017,6.07,0,1.17,1.3,27.23,3.79,0,0,14.72,0,43.22,2.51
Anatolian_BA:I2499:Lazaridis_2017,7.88,0.24,1.61,0,26.13,4.81,0,1.72,13.22,0.03,44.36,0
Anatolian_BA:I2683:Lazaridis_2017,8.73,0.62,2.12,0.42,25.19,0.9,0,0,14.26,0,45.14,2.61
Iberomaurusian:TAF009:Loosdrecht_2018,0,3.54,60.38,2.05,0,0,2.48,18.95,8.83,0,0,3.78
Iberomaurusian:TAF010:Loosdrecht_2018,0,0.55,63.9,3.27,0,0.17,1.39,18.7,5.98,0,0,6.05
Iberomaurusian:TAF011:Loosdrecht_2018,0,1.28,64.11,2.45,0,0.07,0.7,18.22,5.74,0.18,0,7.24
Iberomaurusian:TAF012:Loosdrecht_2018,0,1.79,63.81,0.8,0.44,0,3.27,16.18,5.76,0.59,0,7.35
Iberomaurusian:TAF013:Loosdrecht_2018,0,1.07,61.84,2.72,0,0.02,1.89,18.58,6.99,0,0,6.89
Iberomaurusian:TAF014:Loosdrecht_2018,0,0.74,62.76,2.95,0,0,2.12,18.14,5.34,0,0,7.95
Iberomaurusian:TAF015:Loosdrecht_2018,0,0,62.82,1.87,0,0.9,0,15.76,6.67,3.45,0,8.52
 
Angela said:
The samples labeled Sardinian, which are the HGDP panel chosen by Cavalli Sforza are extremely diverse. Three of the seven have NO Yamnaya and NO Minoan, Anatolian Bronze Age or Ibero-Marusian. He clearly hit the area on the highlands which is most remote but then got a few from surrounding areas which were a little bit more impacted. Still, the Minoans are a small fraction overall and Yamnaya almost non-existent.

So, as I always maintained, on those uplands there are still people who are pretty darn close to the pre-Bronze Age migrations into Europe.

I don't know where ITS or some of the West Sicily samples come from. Even the samples from Behar for West Sicily are very diverse in terms of Anatolia Bronze Age. One might say, ok, perhaps each Sicilian city has a unique population history, but there's variation even within the cities, i.e. the Busby Siracusa and Trapani numbers.

The Central Sicily, East Sicily, and Calabria samples are a bit more uniform, but then they are much fewer in number, so I don't think I can draw a hard and fast conclusion from that.

The Basilicata numbers, from what source I don't know, are all over the place again. It doesn't make sense. It's not even a large province, although it's certainly very mountainous and isolated, the place where even Christ couldn't reach. One would think that drift might have affected uniparentals, but autosomal should, perhaps, be rather uniform.

Campania and Apulia are also very diverse. I can understand it for Apulia, given it's length, but Campania? Unless, perhaps, some samples came from the city and others from the interior.

That's why it's important to know the academic paper used and how the samples were taken.

Except for one sample, the Tuscan HGDP are pretty uniform, and now I do see one sample for TSI, which is probably an average. It would be interesting to see the results for all 30 at some point. There's also a series of Tuscans for whom I don't know the provenance.

The northern samples are pretty uniform.

So, it's the southern samples where we see unexpected and unexplained diversity. We should ask Salento about it when he returns and see where some of these come from.

Meanwhile, I'm going to search through Busby and Behar to see how and where specifically their samples were selected.
Click to expand...

Behar et al 2013 used individuals whose ancestry was self-reported (https://digitalcommons.wayne.edu/cgi/viewcontent.cgi?article=1040&context=humbiol_preprints pag 13, "sample set"), and Hellenthal et al 2015 (Busby was the curator of the Italian samples) used anonymous donours (https://www.science.org/action/downloadSupplement?doi=10.1126/science.1243518&file=hellenthal.sm.pdf pag 67, "Details of the dataset and phasing")
 
Leopoldo Leone said:
Behar et al 2013 used individuals whose ancestry was self-reported (https://digitalcommons.wayne.edu/cgi/viewcontent.cgi?article=1040&context=humbiol_preprints pag 13, "sample set"), and Hellenthal et al 2015 (Busby was the curator of the Italian samples) used anonymous donours (https://www.science.org/action/downloadSupplement?doi=10.1126/science.1243518&file=hellenthal.sm.pdf pag 67, "Details of the dataset and phasing")
Click to expand...


In most cases, the samples collected are self-reported, and geneticists certainly do not check family trees.
 
Pax Augusta said:
Code: 
Bolshoy_Ostrov:3473±87:BOO006:Lamnidis_2018,4.07,44.75,0,0,0,36.53,2.94,0,0,11.3,0,0.41
Yamnaya:RISE240:Allentoft_2015,26.6,1.31,0,0,7.93,64.16,0,0,0,0,0,0
Yamnaya:RISE546:Allentoft_2015,22.07,0,0,0,0,70.45,0,0,0,0,7.49,0
Yamnaya:RISE547:Allentoft_2015,26.54,0,0,0,10.09,63.38,0,0,0,0,0,0
Yamnaya:RISE548:Allentoft_2015,27.76,0,0.93,0,3.91,65.86,0,0,0,0,1.35,0.19
Yamnaya:RISE550:Allentoft_2015,23.64,0.86,0,0,9.61,64.98,0.74,0,0,0,0,0.17
Yamnaya:RISE552:Allentoft_2015,30.94,1.6,0,0.01,2.06,61.11,0.39,0,0,0,3.1,0.78
Yamnaya:I0231_SVP3:Haak_2015,28.48,2.19,0,0,1.43,59.76,2.73,0,0,0,4.53,0.89
Yamnaya:I0357_SVP5:Haak_2015,31.45,0.86,0,0.77,4.05,54.56,2.22,0,0,0,6.09,0
Yamnaya:I0370_SVP10:Haak_2015,24.66,6.11,0,0,0,59.29,0,0,0,0,9.32,0.62
Yamnaya:I0429_SVP38:Haak_2015,26.2,3.3,0,0,2.84,64.62,2.03,0,0,0,0,1
Yamnaya:I0438_SVP50:Haak_2015,23.9,4.62,0,0,0,60.95,1.32,0.31,0,0,8.64,0.26
Yamnaya:I0441_SVP54:Haak_2015,33.48,1.54,0,0,0,56.53,8.46,0,0,0,0,0
Yamnaya:I0443_SVP57:Haak_2015,28.03,2.76,0,0,5.5,59.49,0,0,0,0,3.22,1
Yamnaya:I0444_SVP58:Haak_2015,30.94,0.57,0,0,3.61,59.43,0,0.11,0,0,2.51,2.83
Yamnaya:I2105:Mathieson_2018,25.39,2.85,0,0,1.94,58.76,1.17,0,0,0,8.18,1.72
Yamnaya:I3141:Mathieson_2018,24.66,4.21,0,0,2.47,64.51,1.21,0,0,0,2.93,0
Remedello:RISE487:Allentoft_2015,0,0,1.36,0,75.55,12.12,0,0,0.84,0,9.88,0.24
Remedello:RISE489:Allentoft_2015,0,0,0,0,72.08,12.24,0,0,5.69,0,9.99,0
C_Italian_ChL:R4:Antonio_2019,0,0,2.77,0.12,60.45,3.99,0,0,10.54,0.02,21.81,0.32
C_Italian_ChL:R5:Antonio_2019,0,0,2.76,0,63.45,3.27,0,0,10.15,0,20.2,0.17
C_Italian_ChL:R1014:Antonio_2019,0,0,2.67,0,64.87,0.54,0,0,0,0,31.92,0
C_Italian_N:R2:Antonio_2019,0,0,3.61,0,54.5,0,0,0.17,12.44,0,29.12,0.16
C_Italian_N:R3:Antonio_2019,0,0,4.22,0,52.75,0,0,0,11.28,0.2,31.18,0.38
C_Italian_N:R8:Antonio_2019,0,0,2.42,0,53.24,0,0,0,14.2,0.44,29.46,0.24
C_Italian_N:R9:Antonio_2019,0,0,2.58,0,50.8,0,0,0,13.79,0,32.49,0.34
C_Italian_N:R10:Antonio_2019,0,0,3.35,0,53.37,0,0,0,13.76,0,29.52,0
C_Italian_N:R16:Antonio_2019,0,0,3.41,0,57.73,1.14,0,0,9.12,0.21,28.3,0.09
C_Italian_N:R18:Antonio_2019,0,0,3.72,0,51.41,0,0,0.09,10.24,0,34.54,0
C_Italian_N:R19:Antonio_2019,0,0,4.33,0,55.74,0.85,0,0,9.83,0,29.26,0
Minoan:Lasithi:I0070:Lazaridis_2017,0,0,0.62,0.55,37.53,0,0,0,15.52,0,45.7,0.09
Minoan:Lasithi:I0071:Lazaridis_2017,1.16,0,2.96,0,37.9,0.23,0,0,13.01,0,44.59,0.15
Minoan:Lasithi:I0073:Lazaridis_2017,0,0.05,2.92,0.4,36.38,0,0,0,13.48,0,46.7,0.07
Minoan:Lasithi:I0074:Lazaridis_2017,0.58,0,4.33,0,39.44,0,0,0,12.45,0,43.19,0
Minoan:Lasithi:I9005:Lazaridis_2017,1.52,0,5.68,0,37.33,0,0,0,16.14,0,39.25,0.08
Minoan:Odigitria:I9130:Lazaridis_2017,1.38,0,0.19,0,41.77,0,0,0.28,17.61,0,38.77,0
Minoan:Odigitria:I9131:Lazaridis_2017,5.16,0,0,0,36.32,0,0,0,19.19,0,39.01,0.33
Minoan:Petras_EBA:Pta08:Clemente_2021,0,0,3.85,0.33,34.73,0.38,0.14,0.05,14.68,0.48,43.65,1.72
Anatolian_BA:I2495:Lazaridis_2017,6.07,0,1.17,1.3,27.23,3.79,0,0,14.72,0,43.22,2.51
Anatolian_BA:I2499:Lazaridis_2017,7.88,0.24,1.61,0,26.13,4.81,0,1.72,13.22,0.03,44.36,0
Anatolian_BA:I2683:Lazaridis_2017,8.73,0.62,2.12,0.42,25.19,0.9,0,0,14.26,0,45.14,2.61
Iberomaurusian:TAF009:Loosdrecht_2018,0,3.54,60.38,2.05,0,0,2.48,18.95,8.83,0,0,3.78
Iberomaurusian:TAF010:Loosdrecht_2018,0,0.55,63.9,3.27,0,0.17,1.39,18.7,5.98,0,0,6.05
Iberomaurusian:TAF011:Loosdrecht_2018,0,1.28,64.11,2.45,0,0.07,0.7,18.22,5.74,0.18,0,7.24
Iberomaurusian:TAF012:Loosdrecht_2018,0,1.79,63.81,0.8,0.44,0,3.27,16.18,5.76,0.59,0,7.35
Iberomaurusian:TAF013:Loosdrecht_2018,0,1.07,61.84,2.72,0,0.02,1.89,18.58,6.99,0,0,6.89
Iberomaurusian:TAF014:Loosdrecht_2018,0,0.74,62.76,2.95,0,0,2.12,18.14,5.34,0,0,7.95
Iberomaurusian:TAF015:Loosdrecht_2018,0,0,62.82,1.87,0,0.9,0,15.76,6.67,3.45,0,8.52
Click to expand...
Thanks again for helping. No idea why I bother asking anyone else besides you, Salento, and Palermo Trapani.



Target: Alb_NW6
Distance: 51.1287% / 0.51128687 | R4P
51.2Minoan
27.8Yamnaya
15.8C_Italian_ChL
5.2C_Italian_N



Target: Alb_NW5
Distance: 588.0204% / 5.88020404 | R4P
49.2Minoan
33.0Yamnaya
16.6Remedello
1.2C_Italian_ChL



Target: Alb_NW4
Distance: 198.1709% / 1.98170856 | R4P
38.6Minoan
37.0Yamnaya
18.4Remedello
6.0C_Italian_ChL



Target: Alb_NW3
Distance: 162.9428% / 1.62942833 | R4P
44.6Minoan
29.4Yamnaya
14.6Remedello
11.4C_Italian_ChL



Target: Alb_NW2
Distance: 134.8688% / 1.34868845 | R4P
35.6Minoan
32.6C_Italian_ChL
30.0Yamnaya
1.8Bolshoy_Ostrov



Target: Alb_NW1
Distance: 364.1350% / 3.64135047 | R4P
54.0Minoan
34.2Yamnaya
6.6Remedello
5.2C_Italian_ChL

All 6 from North West Albania (and specifically Catholic) and yet so different. It seems I'm the most Minoan yet one of the most Northern shifted ones.
 
Pax Augusta said:
Sardinian HGDP are from Ogliastra.
Click to expand...


My best recollection, as I said, is that Cavalli-Sforza took samples from Ogliastra and also nearby areas. Chang et al would seem to support that recollection.
"Low effective migration rates separate these provinces from a broad area that extends to the mountainous Gennargentu Massif region, including inland Ogliastra to the west. The Gennargentu region is also where some of the Sardinian individuals in the Human Genome Diversity Project (HGDP) originate (A. Piazza, personal communication).We find the HGDP Sardinia individuals partially overlap with our dataset and include a subset that clusters near the Ogliastra sub-population (Figure S1, S2, Table S1, S2). Thus, we use the term “Gennargentu-region” to describe this ancestry component (red component in Figure 2B). Based on these results, and to simplify analyses going forward, we use individuals from the town of Arzana as a representative of the Gennargentu-region ancestry component."

Given the results of the HGDP samples versus the Arzana and surrounding area samples, it seems that Cavalli Sforza took samples from villages not precisely in the most isolated areas. That would explain why some of the HGDP samples do have a minimum amount of steppe and eastern ancestry, and some have absolutely none.

I do not think we know which 30 TSI individuals (Diekenes writes 21) were chosen by Metspalu to create the subset TSI30. The set originally consisted of over 100 samples (it is the one with at least 3 out of 4 grandparents born in Tuscany). A minority does not appear to be completely Tuscan, both by cross-referencing PCAs and their results. The average of all of them may be considered in line with other Tuscan samples. The other series of Tuscans might be another way of labelling the samples of Murlo, Casentino and Volterra. In some cases it is specified that they are from Murlo or Volterra. But originally there are many more than those released.
Click to expand...

My question was whether there are coordinates for each of the 30 TSI samples, or if there is only one set of coordinates, an average.

Likewise, are the coordinates for the 98 samples of TSI we can access available? Can they be posted?

Who determined that there are non Tuscans in the sample and what was the method used to determine that?


The results change somewhat if averages are used in the model instead of aggregated individuals.



TSI average based on about 98 TSI individuals (model in source with aggregated individuals by Jovialis). 48 out of 98 individuals get Anatolian_BA but they all get higher Central Italy Neolitich and lower Minoan.

kB4XNlJ.png




TSI average based on about 98 TSI individuals (model in source with averages posted by Er Monnezza). Only 7 out of 98 individuals get Anatolian_BA.

S566GJY.png
Click to expand...

Indeed they do.

I never like using averages, so I'll stick with the summary based on the aggregate.

In most cases, the samples collected are self-reported, and geneticists certainly do not check family trees.
Click to expand...

True, but that's still better than using samples, or averages from some unknown person on the internet who could be "adjusting" coordinates to support some agenda or another.


Pax Augusta

quote_icon.png
Originally Posted by Francesco
Do these high Anatolia_BA samples come from the paper by Sarno on Grikos from Southern Italy?



They come from a paper by Lazaridis from 2017, the paper on Mycenaeans and Minoans.​
Click to expand...

Sorry? I see high Anatolia Bronze Age in Behar samples, Busby samples, etc. There are very few samples labeled Lazaridis.

I think we really need to know the source of the samples which don't cite an academic author.
 
Angela said:
My question was whether there are coordinates for each of the 30 TSI samples, or if there is only one set of coordinates, an average. Likewise, are the coordinates for the 98 samples of TSI we can access available? Can they be posted?

Who determined that there are non Tuscans in the sample and what was the method used to determine that?
Click to expand...

There was already a discussion weeks ago by Salento and myself on this.

TSI is the only 1000genomes/HapMap sample where they state that it is based on self-reported individuals with at least 3 out of 4 grandparents born in Tuscany. If they stated this, it is clear that for some of the individuals at least one grandparent was not born in Tuscany. Why else would they have declared this? We are talking about a minority, clearly, the average of all is still within a possible Tuscan average.

Some geneticists, not all, remove outliers from sample sets. They use PCAs and other methods, and keep only those individuals that fall most within the average. In any case, this is statistics, which is why averages are always better than individual results in my opinion.

As can be seen in the PCAs, the TSI cluster is very large, ranging from Piedmont/Liguria to Umbria/Marche. But within this the majority shows a clear trend, where the Tuscany/TSI labels are placed automatically by Vahaduo.

aFjGzma.png


UejNmgQ.png



Angela said:
Indeed they do.
Click to expand...

The fact that some in the first model take some Anatolian_BA and others do not is statistically a problem. In the average it comes out for everyone but for 50% it is 0% actually. In the other model, over 94% get 0% (in fact, only two or three individuals obtain high values).

It should be noted that in the first model Minoan is much lower and Central Italy Neolithic is much higher, and this could be the reason.


Angela said:
I never like using averages, so I'll stick with the summary based on the aggregate.
Click to expand...


These are personal choices; neither path guarantees more accurate results. In general, none of these tools, including those used by academics, are 100 per cent accurate. Calculating the ancestral components of an individual is not like measuring an individual's height or weight. So it is normal that there may be variations in results from model to model. Not to mention that the type of DNA test (the same person testing with more different companies gets different raw data that may give slightly different results), the kind of genotyping, low or high resolution, and so on can affect the results of both private and academic samples.

These are multivariate statistics tools. Not to mention the fact that the creator of nMonte (Vahaduo is simply a platform on which nMonte is implemented) said that he did not guarantee accuracy for results from 5% to 0%. That is why one should stick to general trends in my opinion, and not to the results of individuals, whether private or academic.

With modern samples Dodecad K12b remains one of the best. The problem is that Dodecad K12b does not work very well with some ancestral components based on ancient DNA, such as WHG, and for some populations WHG remains an important information. This is not surprising, because it was created long before ancestral components began to be theorised.


Angela said:
Sorry? I see high Anatolia Bronze Age in Behar samples, Busby samples, etc. There are very few samples labeled Lazaridis.

I think we really need to know the source of the samples which don't cite an academic author.
Click to expand...


These are the samples used in Jovialis' model to represent Anatolia_BA. Jovialis published the reference from the beginning. Had they been published before Lazaridis' 2017 study? Not to my knowledge.

I2495:Lazaridis_2017
I2499:Lazaridis_2017
I2683:Lazaridis_2017

Code: 
Anatolian_BA:I2495:Lazaridis_2017,6.07,0,1.17,1.3,27.23,3.79,0,0,14.72,0,43.22,2.51
Anatolian_BA:I2499:Lazaridis_2017,7.88,0.24,1.61,0,26.13,4.81,0,1.72,13.22,0.03,44.36,0
Anatolian_BA:I2683:Lazaridis_2017,8.73,0.62,2.12,0.42,25.19,0.9,0,0,14.26,0,45.14,2.61

Supplementary information from Lazaridis 2017, p. 32.

 
Last edited:
@Pax,
We'll have to agree to disagree about certain things.

As to the reference to Lazaridis, I wasn't enquiring about which samples were used for Anatolia Bronze Age.

I wanted to know if someone can provide the source for the samples where the academic source isn't named, of which there are many. If it is Salento who provided the data we can ask him when he returns.
 
Pax Augusta said:
Code: 
Bolshoy_Ostrov:3473±87:BOO006:Lamnidis_2018,4.07,44.75,0,0,0,36.53,2.94,0,0,11.3,0,0.41
Yamnaya:RISE240:Allentoft_2015,26.6,1.31,0,0,7.93,64.16,0,0,0,0,0,0
Yamnaya:RISE546:Allentoft_2015,22.07,0,0,0,0,70.45,0,0,0,0,7.49,0
Yamnaya:RISE547:Allentoft_2015,26.54,0,0,0,10.09,63.38,0,0,0,0,0,0
Yamnaya:RISE548:Allentoft_2015,27.76,0,0.93,0,3.91,65.86,0,0,0,0,1.35,0.19
Yamnaya:RISE550:Allentoft_2015,23.64,0.86,0,0,9.61,64.98,0.74,0,0,0,0,0.17
Yamnaya:RISE552:Allentoft_2015,30.94,1.6,0,0.01,2.06,61.11,0.39,0,0,0,3.1,0.78
Yamnaya:I0231_SVP3:Haak_2015,28.48,2.19,0,0,1.43,59.76,2.73,0,0,0,4.53,0.89
Yamnaya:I0357_SVP5:Haak_2015,31.45,0.86,0,0.77,4.05,54.56,2.22,0,0,0,6.09,0
Yamnaya:I0370_SVP10:Haak_2015,24.66,6.11,0,0,0,59.29,0,0,0,0,9.32,0.62
Yamnaya:I0429_SVP38:Haak_2015,26.2,3.3,0,0,2.84,64.62,2.03,0,0,0,0,1
Yamnaya:I0438_SVP50:Haak_2015,23.9,4.62,0,0,0,60.95,1.32,0.31,0,0,8.64,0.26
Yamnaya:I0441_SVP54:Haak_2015,33.48,1.54,0,0,0,56.53,8.46,0,0,0,0,0
Yamnaya:I0443_SVP57:Haak_2015,28.03,2.76,0,0,5.5,59.49,0,0,0,0,3.22,1
Yamnaya:I0444_SVP58:Haak_2015,30.94,0.57,0,0,3.61,59.43,0,0.11,0,0,2.51,2.83
Yamnaya:I2105:Mathieson_2018,25.39,2.85,0,0,1.94,58.76,1.17,0,0,0,8.18,1.72
Yamnaya:I3141:Mathieson_2018,24.66,4.21,0,0,2.47,64.51,1.21,0,0,0,2.93,0
Remedello:RISE487:Allentoft_2015,0,0,1.36,0,75.55,12.12,0,0,0.84,0,9.88,0.24
Remedello:RISE489:Allentoft_2015,0,0,0,0,72.08,12.24,0,0,5.69,0,9.99,0
C_Italian_ChL:R4:Antonio_2019,0,0,2.77,0.12,60.45,3.99,0,0,10.54,0.02,21.81,0.32
C_Italian_ChL:R5:Antonio_2019,0,0,2.76,0,63.45,3.27,0,0,10.15,0,20.2,0.17
C_Italian_ChL:R1014:Antonio_2019,0,0,2.67,0,64.87,0.54,0,0,0,0,31.92,0
C_Italian_N:R2:Antonio_2019,0,0,3.61,0,54.5,0,0,0.17,12.44,0,29.12,0.16
C_Italian_N:R3:Antonio_2019,0,0,4.22,0,52.75,0,0,0,11.28,0.2,31.18,0.38
C_Italian_N:R8:Antonio_2019,0,0,2.42,0,53.24,0,0,0,14.2,0.44,29.46,0.24
C_Italian_N:R9:Antonio_2019,0,0,2.58,0,50.8,0,0,0,13.79,0,32.49,0.34
C_Italian_N:R10:Antonio_2019,0,0,3.35,0,53.37,0,0,0,13.76,0,29.52,0
C_Italian_N:R16:Antonio_2019,0,0,3.41,0,57.73,1.14,0,0,9.12,0.21,28.3,0.09
C_Italian_N:R18:Antonio_2019,0,0,3.72,0,51.41,0,0,0.09,10.24,0,34.54,0
C_Italian_N:R19:Antonio_2019,0,0,4.33,0,55.74,0.85,0,0,9.83,0,29.26,0
Minoan:Lasithi:I0070:Lazaridis_2017,0,0,0.62,0.55,37.53,0,0,0,15.52,0,45.7,0.09
Minoan:Lasithi:I0071:Lazaridis_2017,1.16,0,2.96,0,37.9,0.23,0,0,13.01,0,44.59,0.15
Minoan:Lasithi:I0073:Lazaridis_2017,0,0.05,2.92,0.4,36.38,0,0,0,13.48,0,46.7,0.07
Minoan:Lasithi:I0074:Lazaridis_2017,0.58,0,4.33,0,39.44,0,0,0,12.45,0,43.19,0
Minoan:Lasithi:I9005:Lazaridis_2017,1.52,0,5.68,0,37.33,0,0,0,16.14,0,39.25,0.08
Minoan:Odigitria:I9130:Lazaridis_2017,1.38,0,0.19,0,41.77,0,0,0.28,17.61,0,38.77,0
Minoan:Odigitria:I9131:Lazaridis_2017,5.16,0,0,0,36.32,0,0,0,19.19,0,39.01,0.33
Minoan:Petras_EBA:Pta08:Clemente_2021,0,0,3.85,0.33,34.73,0.38,0.14,0.05,14.68,0.48,43.65,1.72
Anatolian_BA:I2495:Lazaridis_2017,6.07,0,1.17,1.3,27.23,3.79,0,0,14.72,0,43.22,2.51
Anatolian_BA:I2499:Lazaridis_2017,7.88,0.24,1.61,0,26.13,4.81,0,1.72,13.22,0.03,44.36,0
Anatolian_BA:I2683:Lazaridis_2017,8.73,0.62,2.12,0.42,25.19,0.9,0,0,14.26,0,45.14,2.61
Iberomaurusian:TAF009:Loosdrecht_2018,0,3.54,60.38,2.05,0,0,2.48,18.95,8.83,0,0,3.78
Iberomaurusian:TAF010:Loosdrecht_2018,0,0.55,63.9,3.27,0,0.17,1.39,18.7,5.98,0,0,6.05
Iberomaurusian:TAF011:Loosdrecht_2018,0,1.28,64.11,2.45,0,0.07,0.7,18.22,5.74,0.18,0,7.24
Iberomaurusian:TAF012:Loosdrecht_2018,0,1.79,63.81,0.8,0.44,0,3.27,16.18,5.76,0.59,0,7.35
Iberomaurusian:TAF013:Loosdrecht_2018,0,1.07,61.84,2.72,0,0.02,1.89,18.58,6.99,0,0,6.89
Iberomaurusian:TAF014:Loosdrecht_2018,0,0.74,62.76,2.95,0,0,2.12,18.14,5.34,0,0,7.95
Iberomaurusian:TAF015:Loosdrecht_2018,0,0,62.82,1.87,0,0.9,0,15.76,6.67,3.45,0,8.52
Click to expand...

9swguN1.png


Interesting model.
Not sure why the distance is 500% though, lol.
 
You must log in or register to reply here.

This thread has been viewed 69609 times.

Back
Top