Turkish DNA | Turkic and Greek DNA in modern Turkish people.

[Aeolus]

WestByzantine_Greek:Samantaşi,0.09987975,0.150806,-0.0317725,-0.06387325,0.00053875,-0.0237755,-0.001645,-0.005596,-0.00286325,0.0180415,-0.00036575,0.00183575,-0.004571,0.0017205,-0.01316475,0.00487275,0.0150595,0.00079175,0.006002,-0.00719075,-0.00174675,-0.00151475,0.00018475,-0.00506075,-0.00401175

WestByzantine_Greek:Samantaşi,0.110029,0.150637,-0.031678333,-0.060508667,0.0060523333,-0.025658,-0.0029766667,-0.007769,-0.002386,0.022415,0.0099056667,0.0024476667,-0.0058966667,0.0039453333,-0.014024667,-0.00358,0.0069103333,-0.00050666667,0.003394,-0.003877,-0.006738,-0.001113,-0.001315,-0.0030123333,-0.0027543333


Medieval_Turkic:Karakhanid.SGA204_noUDG.SG,0.073985,-0.156392,0.031301,0.013889,-0.052933,0,0.000235,0.003231,-0.014521,-0.017677,-0.014453,-0.003147,0.00223,-0.016377,0.009908,-0.000928,-0.01343,0.001267,0.004525,0.001,-0.014974,0.000742,-0.004067,-0.005663,-0.006586


WestByzantine_Greek:Byzantine_Nicaea:I8366,0.120652,0.155376,-0.029038,-0.067507,0.007694,-0.025658,-0.000235,-0.002077,-0.006136,0.028611,0.013641,0.014387,-0.025867,0.00234,-0.02348,-0.008088,0.014603,0.003167,0.007165,-0.010255,-0.003244,0.009274,-0.003821,0,-0.00491
WestByzantine_Greek:Byzantine_Nicaea:I8368,0.110408,0.152329,-0.03017,-0.071383,0.003385,-0.023706,0.00047,-0.006231,-0.001023,0.028429,-0.002111,-0.000899,-0.010704,0.000413,-0.009365,-0.009414,0.016559,-0.004561,0.006285,-0.010505,-0.010482,0.012242,-0.006655,-0.005784,-0.005628
WestByzantine_Greek:Byzantine_Nicaea:I8370,0.114961,0.161469,-0.028284,-0.068153,0.00277,-0.02761,0.002585,-0.006,-0.011862,0.017312,0.004547,0.002997,-0.016501,0.009083,-0.017644,-0.028109,-0.008736,0.007475,0.015964,-0.008629,-0.005865,0.012736,-0.000246,0.000241,0.002275
WestByzantine_Greek:Byzantine_Nicaea:I8371,0.102441,0.155376,-0.040729,-0.075582,-0.002462,-0.029841,0.004935,-0.009692,-0.009817,0.017312,0.00065,0.004496,-0.003122,0.004679,-0.007465,-0.014452,-0.008084,-0.003674,0.003645,-0.009254,-0.005366,-0.000124,-0.002218,-0.003856,-0.003592
WestByzantine_Greek:Byzantine_Nicaea:I8373,0.112685,0.153345,-0.029038,-0.056525,-0.002462,-0.026216,0.00094,-0.010846,-0.00409,0.029887,0.00682,0.002098,-0.004608,0.003303,-0.020494,-0.020684,-0.012908,0.0019,0.005531,-0.007754,-0.006613,0.00507,0.001602,-0.001084,0.006466
WestByzantine_Greek:Byzantine_Nicaea:I14833,0.110408,0.14319,-0.023381,-0.05814,0.005539,-0.023706,0.002585,-0.008538,-0.008795,0.019317,-0.004709,0.012439,-0.0055,0.007294,-0.019815,-0.009944,-0.009388,-0.001394,0.013827,-0.004752,-0.006863,0.00507,0.007765,0.00735,0.00479
WestByzantine_Greek:Byzantine_Nicaea:I14839,0.100164,0.153345,-0.044123,-0.069445,-0.010463,-0.024542,0.00376,0.001846,-0.003681,0.01221,0.010555,-0.002847,0.000149,0.003165,-0.015608,0.01432,0.014473,0.003674,0.005782,-0.006378,-0.001373,0.005812,-0.000616,0.006748,-0.005269
WestByzantine_Greek:Byzantine_Nicaea:I14842,0.114961,0.15436,-0.028284,-0.062985,0.003385,-0.02761,-0.00094,-0.002308,-0.00225,0.034078,-0.000162,0.00045,-0.01115,0.005918,-0.020358,-0.0179,-0.016037,0,0.006788,-0.004627,-0.00574,0.000989,-0.012818,-0.007109,-0.006586
WestByzantine_Greek:Byzantine_Nicaea:I14843,0.094473,0.149283,-0.024136,-0.071383,0.005847,-0.027331,-0.005405,0.001154,-0.000205,0.025878,0.008282,0.003447,-0.007136,-0.004954,-0.014929,0.007425,0.023469,-0.001774,0.007919,-0.005378,-0.006863,0.00371,0.001479,0.007471,-0.002634
WestByzantine_Greek:Byzantine_Balikesir:I14821,0.114961,0.141159,-0.016216,-0.037145,0.007078,-0.01255,0.00611,-0.003923,-0.006136,0.007107,0.004709,0.003147,-0.004014,0.002202,-0.005972,-0.000265,0.012778,0.008108,0.006662,-0.000875,0.006364,-0.003091,-0.001479,-0.005904,0.011855
WestByzantine_Greek:Byzantine_Balikesir:I14822,0.105855,0.142174,-0.004525,-0.02584,0.015387,-0.006693,0.007755,-0.008769,-0.00225,0.010023,0.000812,0.007493,-0.007433,0.010184,-0.005293,-0.009016,-0.0103,0.000253,0.005531,-0.007879,-0.000998,-0.007048,-0.002095,0.001687,-0.008143
WestByzantine_Greek:Byzantine_Balikesir:I14831,0.119514,0.148267,-0.007165,-0.047158,0.014464,-0.016176,0.00517,0.006461,-0.005522,0.011299,-0.002598,0.000899,-0.00446,0.00578,-0.009772,-0.014585,-0.009257,0.006208,0.009302,-0.015632,0.000873,-0.002597,-0.003451,-0.00241,-0.005149
WestByzantine_Greek:Byzantine_Balikesir:I14832,0.111547,0.14319,0.001886,-0.037468,0.01231,-0.01004,0.007755,0.001154,-0.009204,0.009476,0.004709,0.000599,0.004608,0.003441,-0.002579,-0.013259,-0.007953,0.00076,0.003645,-0.004877,-0.005366,0,0.009613,-0.000482,-0.004191


WestByzantine_Greek:StratonikeiaAVG,0.10341657,0.14551086,-0.031193286,-0.059755,-0.00013185714,-0.021713714,-0.0013092857,-0.0034285714,-0.0070708571,0.014865143,0.002227,0.0046242857,-0.0051604286,0.0010814286,-0.012195429,-0.00053014286,0.0085495714,0.00061542857,0.00501,-0.00098257143,-0.0041532857,0.002014,-0.000035142857,0.0016697143,-0.0042254286

This is for western Turks. I actually listed all samples anyway with their names, so just look them up on illustrativedna.com if you mean a specific one.

Could you share all you used, so also for other parts of Anatolia? I'm curious. Btw, why are you even using that many different Byzantine era pops for different regions of Anatolia?

Btw what even is Turkey_West that you used? The distances I'm getting are barely passable (and into the zone of invalid imo). Sorry about the formatting:

Target,                    Distance,        WestByzantine_Greek,    Medieval_Turkic
Turkish_Northwest, 0.01505590,   70.4,                             29.6
Turkish_Southwest, 0.01501872,   70.0,                             30.0
Average,                  0.01503731,   70.2,                            29.8

And here is one with a Colchis-like sample in the model (I'm not suggesting there is Caucasian-admix necessarily, rather something generally eastern, just that only Byzantine Greek doesn't seem enough):

Target                 Distance         WestByzantine_Greek Medieval_Turkic       Colchis-like_TUR_BlackSea_Samsun_Ancient_B
Turkish_Northwest 0.00989572 60.6                         28.8                       10.6
Turkish_Southwest 0.01189951 62.4                         29.2                       8.4
Average                 0.01089761 61.5                         29.0                       9.5

See the massive difference in distance. Or with Hasanlu_IA:

Target                 Distance         WestByzantine_Greek Medieval_Turkic   IRN_Hasanlu_IA
Turkish_Northwest 0.01271794 59.2                         28.0                   12.8
Turkish_Southwest 0.01158726 54.0                         27.6                   18.4
Average                 0.01215260 56.6                         27.8                   15.6

There aren't. The samples are all from Illustrative DNA. We don't even have caucasian admixed Muğla Greek samples, so idk what you're doing.

[Aeolus]

WestByzantine_Greek:Samantaşi,0.09987975,0.150806,-0.0317725,-0.06387325,0.00053875,-0.0237755,-0.001645,-0.005596,-0.00286325,0.0180415,-0.00036575,0.00183575,-0.004571,0.0017205,-0.01316475,0.00487275,0.0150595,0.00079175,0.006002,-0.00719075,-0.00174675,-0.00151475,0.00018475,-0.00506075,-0.00401175

WestByzantine_Greek:Samantaşi,0.110029,0.150637,-0.031678333,-0.060508667,0.0060523333,-0.025658,-0.0029766667,-0.007769,-0.002386,0.022415,0.0099056667,0.0024476667,-0.0058966667,0.0039453333,-0.014024667,-0.00358,0.0069103333,-0.00050666667,0.003394,-0.003877,-0.006738,-0.001113,-0.001315,-0.0030123333,-0.0027543333


Medieval_Turkic:Karakhanid.SGA204_noUDG.SG,0.073985,-0.156392,0.031301,0.013889,-0.052933,0,0.000235,0.003231,-0.014521,-0.017677,-0.014453,-0.003147,0.00223,-0.016377,0.009908,-0.000928,-0.01343,0.001267,0.004525,0.001,-0.014974,0.000742,-0.004067,-0.005663,-0.006586


WestByzantine_Greek:Byzantine_Nicaea:I8366,0.120652,0.155376,-0.029038,-0.067507,0.007694,-0.025658,-0.000235,-0.002077,-0.006136,0.028611,0.013641,0.014387,-0.025867,0.00234,-0.02348,-0.008088,0.014603,0.003167,0.007165,-0.010255,-0.003244,0.009274,-0.003821,0,-0.00491
WestByzantine_Greek:Byzantine_Nicaea:I8368,0.110408,0.152329,-0.03017,-0.071383,0.003385,-0.023706,0.00047,-0.006231,-0.001023,0.028429,-0.002111,-0.000899,-0.010704,0.000413,-0.009365,-0.009414,0.016559,-0.004561,0.006285,-0.010505,-0.010482,0.012242,-0.006655,-0.005784,-0.005628
WestByzantine_Greek:Byzantine_Nicaea:I8370,0.114961,0.161469,-0.028284,-0.068153,0.00277,-0.02761,0.002585,-0.006,-0.011862,0.017312,0.004547,0.002997,-0.016501,0.009083,-0.017644,-0.028109,-0.008736,0.007475,0.015964,-0.008629,-0.005865,0.012736,-0.000246,0.000241,0.002275
WestByzantine_Greek:Byzantine_Nicaea:I8371,0.102441,0.155376,-0.040729,-0.075582,-0.002462,-0.029841,0.004935,-0.009692,-0.009817,0.017312,0.00065,0.004496,-0.003122,0.004679,-0.007465,-0.014452,-0.008084,-0.003674,0.003645,-0.009254,-0.005366,-0.000124,-0.002218,-0.003856,-0.003592
WestByzantine_Greek:Byzantine_Nicaea:I8373,0.112685,0.153345,-0.029038,-0.056525,-0.002462,-0.026216,0.00094,-0.010846,-0.00409,0.029887,0.00682,0.002098,-0.004608,0.003303,-0.020494,-0.020684,-0.012908,0.0019,0.005531,-0.007754,-0.006613,0.00507,0.001602,-0.001084,0.006466
WestByzantine_Greek:Byzantine_Nicaea:I14833,0.110408,0.14319,-0.023381,-0.05814,0.005539,-0.023706,0.002585,-0.008538,-0.008795,0.019317,-0.004709,0.012439,-0.0055,0.007294,-0.019815,-0.009944,-0.009388,-0.001394,0.013827,-0.004752,-0.006863,0.00507,0.007765,0.00735,0.00479
WestByzantine_Greek:Byzantine_Nicaea:I14839,0.100164,0.153345,-0.044123,-0.069445,-0.010463,-0.024542,0.00376,0.001846,-0.003681,0.01221,0.010555,-0.002847,0.000149,0.003165,-0.015608,0.01432,0.014473,0.003674,0.005782,-0.006378,-0.001373,0.005812,-0.000616,0.006748,-0.005269
WestByzantine_Greek:Byzantine_Nicaea:I14842,0.114961,0.15436,-0.028284,-0.062985,0.003385,-0.02761,-0.00094,-0.002308,-0.00225,0.034078,-0.000162,0.00045,-0.01115,0.005918,-0.020358,-0.0179,-0.016037,0,0.006788,-0.004627,-0.00574,0.000989,-0.012818,-0.007109,-0.006586
WestByzantine_Greek:Byzantine_Nicaea:I14843,0.094473,0.149283,-0.024136,-0.071383,0.005847,-0.027331,-0.005405,0.001154,-0.000205,0.025878,0.008282,0.003447,-0.007136,-0.004954,-0.014929,0.007425,0.023469,-0.001774,0.007919,-0.005378,-0.006863,0.00371,0.001479,0.007471,-0.002634
WestByzantine_Greek:Byzantine_Balikesir:I14821,0.114961,0.141159,-0.016216,-0.037145,0.007078,-0.01255,0.00611,-0.003923,-0.006136,0.007107,0.004709,0.003147,-0.004014,0.002202,-0.005972,-0.000265,0.012778,0.008108,0.006662,-0.000875,0.006364,-0.003091,-0.001479,-0.005904,0.011855
WestByzantine_Greek:Byzantine_Balikesir:I14822,0.105855,0.142174,-0.004525,-0.02584,0.015387,-0.006693,0.007755,-0.008769,-0.00225,0.010023,0.000812,0.007493,-0.007433,0.010184,-0.005293,-0.009016,-0.0103,0.000253,0.005531,-0.007879,-0.000998,-0.007048,-0.002095,0.001687,-0.008143
WestByzantine_Greek:Byzantine_Balikesir:I14831,0.119514,0.148267,-0.007165,-0.047158,0.014464,-0.016176,0.00517,0.006461,-0.005522,0.011299,-0.002598,0.000899,-0.00446,0.00578,-0.009772,-0.014585,-0.009257,0.006208,0.009302,-0.015632,0.000873,-0.002597,-0.003451,-0.00241,-0.005149
WestByzantine_Greek:Byzantine_Balikesir:I14832,0.111547,0.14319,0.001886,-0.037468,0.01231,-0.01004,0.007755,0.001154,-0.009204,0.009476,0.004709,0.000599,0.004608,0.003441,-0.002579,-0.013259,-0.007953,0.00076,0.003645,-0.004877,-0.005366,0,0.009613,-0.000482,-0.004191


WestByzantine_Greek:StratonikeiaAVG,0.10341657,0.14551086,-0.031193286,-0.059755,-0.00013185714,-0.021713714,-0.0013092857,-0.0034285714,-0.0070708571,0.014865143,0.002227,0.0046242857,-0.0051604286,0.0010814286,-0.012195429,-0.00053014286,0.0085495714,0.00061542857,0.00501,-0.00098257143,-0.0041532857,0.002014,-0.000035142857,0.0016697143,-0.0042254286

This is for western Turks. I actually listed all samples anyway with their names, so just look them up on illustrativedna.com if you mean a specific one.

Could you share all you used, so also for other parts of Anatolia? I'm curious. Btw, why are you even using that many different Byzantine era pops for different regions of Anatolia?

Btw what even is Turkey_West that you used? The distances I'm getting are barely passable (and into the zone of invalid imo). Sorry about the formatting:

Target,                    Distance,        WestByzantine_Greek,    Medieval_Turkic
Turkish_Northwest, 0.01505590,   70.4,                             29.6
Turkish_Southwest, 0.01501872,   70.0,                             30.0
Average,                  0.01503731,   70.2,                            29.8

And here is one with a Colchis-like sample in the model (I'm not suggesting there is Caucasian-admix necessarily, rather something generally eastern, just that only Byzantine Greek doesn't seem enough):

Target                 Distance         WestByzantine_Greek Medieval_Turkic       Colchis-like_TUR_BlackSea_Samsun_Ancient_B
Turkish_Northwest 0.00989572 60.6                         28.8                       10.6
Turkish_Southwest 0.01189951 62.4                         29.2                       8.4
Average                 0.01089761 61.5                         29.0                       9.5

See the massive difference in distance. Or with Hasanlu_IA:

Target                 Distance         WestByzantine_Greek Medieval_Turkic   IRN_Hasanlu_IA
Turkish_Northwest 0.01271794 59.2                         28.0                   12.8
Turkish_Southwest 0.01158726 54.0                         27.6                   18.4
Average                 0.01215260 56.6                         27.8                   15.6

Send me that Samsun sample if you don't mind. That's probably the Carian-like sample from illustrativedna (Roman Amisos)

[BowFX]

Could you share all you used, so also for other parts of Anatolia? I'm curious. Btw, why are you even using that many different Byzantine era pops for different regions of Anatolia?

Btw what even is Turkey_West that you used? The distances I'm getting are barely passable (and into the zone of invalid imo). Sorry about the formatting:

Target,                    Distance,        WestByzantine_Greek,    Medieval_Turkic
Turkish_Northwest, 0.01505590,   70.4,                             29.6
Turkish_Southwest, 0.01501872,   70.0,                             30.0
Average,                  0.01503731,   70.2,                            29.8

And here is one with a Colchis-like sample in the model (I'm not suggesting there is Caucasian-admix necessarily, rather something generally eastern, just that only Byzantine Greek doesn't seem enough):

Target                 Distance         WestByzantine_Greek Medieval_Turkic       Colchis-like_TUR_BlackSea_Samsun_Ancient_B
Turkish_Northwest 0.00989572 60.6                         28.8                       10.6
Turkish_Southwest 0.01189951 62.4                         29.2                       8.4
Average                 0.01089761 61.5                         29.0                       9.5

See the massive difference in distance. Or with Hasanlu_IA:

Target                 Distance         WestByzantine_Greek Medieval_Turkic   IRN_Hasanlu_IA
Turkish_Northwest 0.01271794 59.2                         28.0                   12.8
Turkish_Southwest 0.01158726 54.0                         27.6                   18.4
Average                 0.01215260 56.6                         27.8                   15.6

Send me that Samsun sample if you don't mind. That's probably the Carian-like sample from illustrativedna (Roman Amisos)

Nah, Samsun_B is very much Caucasian-like. It's Samsun_A that is Anatolian-like. There is also Samsun_C which seems admixed with Steppe Scythians/Cimmerians/Sarmatians/etc., but I do not have quick access to that currently.

Anatolian-like: TUR_BlackSea_Samsun_Ancient_A,0.109839,0.146744,-0.0386545,-0.0686375,0.002,-0.022172,0.0022325,-0.0039225,-0.0139075,0.021048,0.0060895,0.0051705,-0.000595,0.000688,-0.0172365,-0.0017235,0.00515,-0.00038,0.005468,-0.005753,0.000936,0,0.0001235,0.001205,0.001736

Caucasian-like: TUR_BlackSea_Samsun_Ancient_B,0.106994,0.131003,-0.065242,-0.029716,-0.047393,-0.000558,0.004465,-0.005307,-0.06463,-0.027335,-0.001624,0.012739,-0.026908,0.00812,0.003257,-0.032617,0.005867,-0.013682,-0.005154,0.016133,0.013601,0.004081,0.004314,-0.016147,-0.007784

Btw do keep in mind that these are averages (from Southern Arc in fact), but iirc each is made up of just a few singles. Samsun_B might even be just one or two people, but I absolutely can't guarantee that.

Also, what even is Roman Amisos? Could you share the coordinates?

[BowFX]

Could you share all you used, so also for other parts of Anatolia? I'm curious. Btw, why are you even using that many different Byzantine era pops for different regions of Anatolia?

Btw what even is Turkey_West that you used? The distances I'm getting are barely passable (and into the zone of invalid imo). Sorry about the formatting:

Target,                    Distance,        WestByzantine_Greek,    Medieval_Turkic
Turkish_Northwest, 0.01505590,   70.4,                             29.6
Turkish_Southwest, 0.01501872,   70.0,                             30.0
Average,                  0.01503731,   70.2,                            29.8

And here is one with a Colchis-like sample in the model (I'm not suggesting there is Caucasian-admix necessarily, rather something generally eastern, just that only Byzantine Greek doesn't seem enough):

Target                 Distance         WestByzantine_Greek Medieval_Turkic       Colchis-like_TUR_BlackSea_Samsun_Ancient_B
Turkish_Northwest 0.00989572 60.6                         28.8                       10.6
Turkish_Southwest 0.01189951 62.4                         29.2                       8.4
Average                 0.01089761 61.5                         29.0                       9.5

See the massive difference in distance. Or with Hasanlu_IA:

Target                 Distance         WestByzantine_Greek Medieval_Turkic   IRN_Hasanlu_IA
Turkish_Northwest 0.01271794 59.2                         28.0                   12.8
Turkish_Southwest 0.01158726 54.0                         27.6                   18.4
Average                 0.01215260 56.6                         27.8                   15.6

There aren't. The samples are all from Illustrative DNA. We don't even have caucasian admixed Muğla Greek samples, so idk what you're doing.

Then why do the distances improve so much when Samsun_B or IRN_Hasanlu_IA is added? This 'eastern' admix might have come during or shortly before the Turkic migrations. Thought maybe G25 is just tripping, I wouldn't know.

[Aeolus]

Btw what even is Turkey_West that you used? The distances I'm getting are barely passable (and into the zone of invalid imo). Sorry about the formatting:

Target,                    Distance,        WestByzantine_Greek,    Medieval_Turkic
Turkish_Northwest, 0.01505590,   70.4,                             29.6
Turkish_Southwest, 0.01501872,   70.0,                             30.0
Average,                  0.01503731,   70.2,                            29.8

And here is one with a Colchis-like sample in the model (I'm not suggesting there is Caucasian-admix necessarily, rather something generally eastern, just that only Byzantine Greek doesn't seem enough):

Target                 Distance         WestByzantine_Greek Medieval_Turkic       Colchis-like_TUR_BlackSea_Samsun_Ancient_B
Turkish_Northwest 0.00989572 60.6                         28.8                       10.6
Turkish_Southwest 0.01189951 62.4                         29.2                       8.4
Average                 0.01089761 61.5                         29.0                       9.5

See the massive difference in distance. Or with Hasanlu_IA:

Target                 Distance         WestByzantine_Greek Medieval_Turkic   IRN_Hasanlu_IA
Turkish_Northwest 0.01271794 59.2                         28.0                   12.8
Turkish_Southwest 0.01158726 54.0                         27.6                   18.4
Average                 0.01215260 56.6                         27.8                   15.6

There aren't. The samples are all from Illustrative DNA. We don't even have caucasian admixed Muğla Greek samples, so idk what you're doing.

Then why do the distances improve so much when Samsun_B or IRN_Hasanlu_IA is added? This 'eastern' admix might have come during or shortly before the Turkic migrations. Thought maybe G25 is just tripping, I wouldn't know.

If using an Eastern admix, it obviously overfits. Turkomans were half Iranian, but that doesn't mean every western Anatolian Turk is half Iranian as well. Eastern Turks descent from Turkomans, we don't. 

Since Turks are already a mixture of 2 different steppe sources, multiple eastern Mediterranean sources, and on top of that have varying degrees of East Asian admixture. Turks can be easily modelled as anything, really.

Ottoman Menteshe is 15% Iranian, due to some Turks being half Turkoman (1/4th Iranian). There's no Iranian admixture or caucasian admixture in Western Anatolian Turks. I don't know why you obsess about it so much when even Qpadm doesn't show it.

   
   
   

Oh, look, we're actually all half Armenian because the fit is better xD

   

I already explained above why it's an overfit.

Here are Ottoman Turks, for example. They're 15% Iranian. Difference between Ottoman Turks and modern Turks is that modern Turks lack the Iranian admixture and have a few % lesser Turkic and thus more Greek.

   

[Aeolus]

Btw what even is Turkey_West that you used? The distances I'm getting are barely passable (and into the zone of invalid imo). Sorry about the formatting:

Target,                    Distance,        WestByzantine_Greek,    Medieval_Turkic
Turkish_Northwest, 0.01505590,   70.4,                             29.6
Turkish_Southwest, 0.01501872,   70.0,                             30.0
Average,                  0.01503731,   70.2,                            29.8

And here is one with a Colchis-like sample in the model (I'm not suggesting there is Caucasian-admix necessarily, rather something generally eastern, just that only Byzantine Greek doesn't seem enough):

Target                 Distance         WestByzantine_Greek Medieval_Turkic       Colchis-like_TUR_BlackSea_Samsun_Ancient_B
Turkish_Northwest 0.00989572 60.6                         28.8                       10.6
Turkish_Southwest 0.01189951 62.4                         29.2                       8.4
Average                 0.01089761 61.5                         29.0                       9.5

See the massive difference in distance. Or with Hasanlu_IA:

Target                 Distance         WestByzantine_Greek Medieval_Turkic   IRN_Hasanlu_IA
Turkish_Northwest 0.01271794 59.2                         28.0                   12.8
Turkish_Southwest 0.01158726 54.0                         27.6                   18.4
Average                 0.01215260 56.6                         27.8                   15.6

Send me that Samsun sample if you don't mind. That's probably the Carian-like sample from illustrativedna (Roman Amisos)

Nah, Samsun_B is very much Caucasian-like. It's Samsun_A that is Anatolian-like. There is also Samsun_C which seems admixed with Steppe Scythians/Cimmerians/Sarmatians/etc., but I do not have quick access to that currently.

Anatolian-like: TUR_BlackSea_Samsun_Ancient_A,0.109839,0.146744,-0.0386545,-0.0686375,0.002,-0.022172,0.0022325,-0.0039225,-0.0139075,0.021048,0.0060895,0.0051705,-0.000595,0.000688,-0.0172365,-0.0017235,0.00515,-0.00038,0.005468,-0.005753,0.000936,0,0.0001235,0.001205,0.001736

Caucasian-like: TUR_BlackSea_Samsun_Ancient_B,0.106994,0.131003,-0.065242,-0.029716,-0.047393,-0.000558,0.004465,-0.005307,-0.06463,-0.027335,-0.001624,0.012739,-0.026908,0.00812,0.003257,-0.032617,0.005867,-0.013682,-0.005154,0.016133,0.013601,0.004081,0.004314,-0.016147,-0.007784

Btw do keep in mind that these are averages (from Southern Arc in fact), but iirc each is made up of just a few singles. Samsun_B might even be just one or two people, but I absolutely can't guarantee that.

Also, what even is Roman Amisos? Could you share the coordinates?

Roman_Samsun:Amisos:I4530,0.117238,0.147252,-0.034695,-0.066861,0.007078,-0.024263,0.00329,-0.006,-0.008385,0.026424,0.00341,0.005395,-0.004906,-0.002064,-0.016694,0.009149,0.01369,0.0019,0.000251,-5.00E-04,0.000873,-0.002349,-0.002835,-0.000241,-0.0

Roman_Samsun:Amisos:I4530,0.117238,0.147252,-0.034695,-0.066861,0.007078,-0.024263,0.00329,-0.006,-0.008385,0.026424,0.00341,0.005395,-0.004906,-0.002064,-0.016694,0.009149,0.01369,0.0019,0.000251,-5.00E-04,0.000873,-0.002349,-0.002835,-0.000241,-0.000958

It's Roman Era Samsun and plots with Carians and Southwest Byzantines.

[BowFX]

There aren't. The samples are all from Illustrative DNA. We don't even have caucasian admixed Muğla Greek samples, so idk what you're doing.

Then why do the distances improve so much when Samsun_B or IRN_Hasanlu_IA is added? This 'eastern' admix might have come during or shortly before the Turkic migrations. Thought maybe G25 is just tripping, I wouldn't know.

If using an Eastern admix, it obviously overfits. Turkomans were half Iranian, but that doesn't mean every western Anatolian Turk is half Iranian as well. Eastern Turks descent from Turkomans, we don't. 

Since Turks are already a mixture of 2 different steppe sources, multiple eastern Mediterranean sources, and on top of that have varying degrees of East Asian admixture. Turks can be easily modelled as anything, really.

Ottoman Menteshe is 15% Iranian, due to some Turks being half Turkoman (1/4th Iranian). There's no Iranian admixture or caucasian admixture in Western Anatolian Turks. I don't know why you obsess about it so much when even Qpadm doesn't show it.





Oh, look, we're actually all half Armenian because the fit is better xD



I already explained above why it's an overfit.

Here are Ottoman Turks, for example. They're 15% Iranian. Difference between Ottoman Turks and modern Turks is that modern Turks lack the Iranian admixture and have a few % lesser Turkic and thus more Greek.

Which two Steppe sources do you mean?

Also, I'm not obsessed with eastern admix, it's just that the distances are too high without it, which is why I'm wondering whether there is something funky going on with G25.

And why would Ottoman Turks have Iranian but not modern Turks? Just in case it's not obvious on the internet, I'm not arguing, I genuinely just want to hear why you think so.

[BowFX]

Send me that Samsun sample if you don't mind. That's probably the Carian-like sample from illustrativedna (Roman Amisos)

Nah, Samsun_B is very much Caucasian-like. It's Samsun_A that is Anatolian-like. There is also Samsun_C which seems admixed with Steppe Scythians/Cimmerians/Sarmatians/etc., but I do not have quick access to that currently.

Anatolian-like: TUR_BlackSea_Samsun_Ancient_A,0.109839,0.146744,-0.0386545,-0.0686375,0.002,-0.022172,0.0022325,-0.0039225,-0.0139075,0.021048,0.0060895,0.0051705,-0.000595,0.000688,-0.0172365,-0.0017235,0.00515,-0.00038,0.005468,-0.005753,0.000936,0,0.0001235,0.001205,0.001736

Caucasian-like: TUR_BlackSea_Samsun_Ancient_B,0.106994,0.131003,-0.065242,-0.029716,-0.047393,-0.000558,0.004465,-0.005307,-0.06463,-0.027335,-0.001624,0.012739,-0.026908,0.00812,0.003257,-0.032617,0.005867,-0.013682,-0.005154,0.016133,0.013601,0.004081,0.004314,-0.016147,-0.007784

Btw do keep in mind that these are averages (from Southern Arc in fact), but iirc each is made up of just a few singles. Samsun_B might even be just one or two people, but I absolutely can't guarantee that.

Also, what even is Roman Amisos? Could you share the coordinates?

Roman_Samsun:Amisos:I4530,0.117238,0.147252,-0.034695,-0.066861,0.007078,-0.024263,0.00329,-0.006,-0.008385,0.026424,0.00341,0.005395,-0.004906,-0.002064,-0.016694,0.009149,0.01369,0.0019,0.000251,-5.00E-04,0.000873,-0.002349,-0.002835,-0.000241,-0.0

Roman_Samsun:Amisos:I4530,0.117238,0.147252,-0.034695,-0.066861,0.007078,-0.024263,0.00329,-0.006,-0.008385,0.026424,0.00341,0.005395,-0.004906,-0.002064,-0.016694,0.009149,0.01369,0.0019,0.000251,-5.00E-04,0.000873,-0.002349,-0.002835,-0.000241,-0.000958

It's Roman Era Samsun and plots with Carians and Southwest Byzantines.

I just checked the distances, that sample is indeed close to Samsun_A, so also close to BA Anatolian and Southwest Byzantines, but it has basically nothing to do with Samsun_B. Samsun_B is very similar to Georgian (when modelling Anatolians at least).

Btw I saw your model in your signature. Where do you have coords from? Because that distance is insanely low for real individuals imo, among the lowest I've seen.

[Aeolus]

Then why do the distances improve so much when Samsun_B or IRN_Hasanlu_IA is added? This 'eastern' admix might have come during or shortly before the Turkic migrations. Thought maybe G25 is just tripping, I wouldn't know.

If using an Eastern admix, it obviously overfits. Turkomans were half Iranian, but that doesn't mean every western Anatolian Turk is half Iranian as well. Eastern Turks descent from Turkomans, we don't. 

Since Turks are already a mixture of 2 different steppe sources, multiple eastern Mediterranean sources, and on top of that have varying degrees of East Asian admixture. Turks can be easily modelled as anything, really.

Ottoman Menteshe is 15% Iranian, due to some Turks being half Turkoman (1/4th Iranian). There's no Iranian admixture or caucasian admixture in Western Anatolian Turks. I don't know why you obsess about it so much when even Qpadm doesn't show it.





Oh, look, we're actually all half Armenian because the fit is better xD



I already explained above why it's an overfit.

Here are Ottoman Turks, for example. They're 15% Iranian. Difference between Ottoman Turks and modern Turks is that modern Turks lack the Iranian admixture and have a few % lesser Turkic and thus more Greek.

Which two Steppe sources do you mean?

Also, I'm not obsessed with eastern admix, it's just that the distances are too high without it, which is why I'm wondering whether there is something funky going on with G25.

And why would Ottoman Turks have Iranian but not modern Turks? Just in case it's not obvious on the internet, I'm not arguing, I genuinely just want to hear why you think so.

It's not "way too high". 1.2 is a perfect fit. Anything below 1 is an overfit anyways. Why do you try to model Turks as if they're a homogeneous ethnic group? 

The 2 Steppe sources: One is from the Turkic source and the other from the pre-turkic source.

[BowFX]

If using an Eastern admix, it obviously overfits. Turkomans were half Iranian, but that doesn't mean every western Anatolian Turk is half Iranian as well. Eastern Turks descent from Turkomans, we don't. 

Since Turks are already a mixture of 2 different steppe sources, multiple eastern Mediterranean sources, and on top of that have varying degrees of East Asian admixture. Turks can be easily modelled as anything, really.

Ottoman Menteshe is 15% Iranian, due to some Turks being half Turkoman (1/4th Iranian). There's no Iranian admixture or caucasian admixture in Western Anatolian Turks. I don't know why you obsess about it so much when even Qpadm doesn't show it.





Oh, look, we're actually all half Armenian because the fit is better xD



I already explained above why it's an overfit.

Here are Ottoman Turks, for example. They're 15% Iranian. Difference between Ottoman Turks and modern Turks is that modern Turks lack the Iranian admixture and have a few % lesser Turkic and thus more Greek.

Which two Steppe sources do you mean?

Also, I'm not obsessed with eastern admix, it's just that the distances are too high without it, which is why I'm wondering whether there is something funky going on with G25.

And why would Ottoman Turks have Iranian but not modern Turks? Just in case it's not obvious on the internet, I'm not arguing, I genuinely just want to hear why you think so.

It's not "way too high". 1.2 is a perfect fit. Anything below 1 is an overfit anyways. Why do you try to model Turks as if they're a homogeneous ethnic group? 

The 2 Steppe sources: One is from the Turkic source and the other from the pre-turkic source.

Umm, Turkish_Northwest and Southwest from the G25 list get a fit of 1.5, at which point it looks like something is missing. I don't have the coords of Turkey_West that you are using. As for why I model Turks as if they're a homogenous ethnic group, you obviously have to draw the line somewhere, can't use a different model for each and every subregion.

We obviously need Byzantine-era samples from all over, but for now I'll assume there was a "core East-Med" in most parts of Anatolia (and by proxy Greece, Balkans, Cyprus, etc.), with each region possibly showing a shift, which can be captured by non-Byzantine pops. This doesn't mean those regions have exactly that much admix from said pops, they are just there for comparison.

Again, not arguing.