Pretty cool results.
Toppo, I will get with you to get my G25 results. I can pay tomorrow. I am curious on mines.
Printable View
Pretty cool results.
Toppo, I will get with you to get my G25 results. I can pay tomorrow. I am curious on mines.
Well, he only gave me the coordinates.Quote:
Originally Posted by Mingle
Cool. I have to figure out how to get the coordinates.Quote:
Originally Posted by Toppo900
Really? I mean, isn't fits that are below 1 considered to be very accurate?Quote:
Originally Posted by Mingle
I used unscaled coordinates using only modern samples for Palestinian Muslims:
"sample": "Test1:Palestinian",
"fit": 0.7489,
"Lebanese_Christian": 63.33,
"Egyptian": 17.5,
"BedouinB": 15,
"Dinka": 4.17,
Isn't the results accurate? Here's using ancient samples:
"sample": "Test1:Palestinian",
"fit": 0.761,
"Levant_Canaanite_MBA": 58.33,
"Levant_JOR_EBA": 19.17,
"IRN_Hajji_Firuz_BA": 13.33,
"Canary_Islands_Guanche": 5,
"TZA_Pemba_600BP": 4.17,
Here's the results for the Palestinian Muslims using scaled ancient coordinates:
"sample": "Test1:Palestinian",
"fit": 1.5156,
"Levant_Canaanite_MBA": 56.67,
"Levant_JOR_EBA": 17.5,
"Canary_Islands_Guanche": 10.83,
"TKM_IA": 6.67,
"IRN_Hajji_Firuz_BA": 5.83,
"TZA_Pemba_600BP": 2.5,
If you use overlapping components, there will definitely be overfitting. You get lower fits, but the result is overfitted. For example, to model SW Asians like Palestinians using ancients, I would use Levant_N, Anatolia_N, Iran_Neolithic, CHG, and Iberomaurisian. There's some similarities between Anatolia_N and Levant_N. But these are less overlapping because Anatolia_N is significantly drifted towards WHG compared to Levant_N. I guess you could use similar idea to model with bronze age and iron age pops maybe?Quote:
Originally Posted by Toppo900
What's levant_N? Do you mean Levantine Natufian or Canaanite from Syria?Quote:
Originally Posted by NPKTO
It was Levantine Neolithic population. You could also replace it with Levant_PPNB (Pre Pottery Neolithic Culture B).Quote:
Originally Posted by Toppo900
What about CHG since it's not found in the spreadsheets?Quote:
Originally Posted by NPKTO
If you use ancient components and you get a low fit, I'd say its pretty good. Overfitting is when genetically similar components steal from each other.Quote:
Originally Posted by Toppo900
I see. So, over-fitting is bad when using ancient samples but it's good when using modern samples like in the first results, right?Quote:
Originally Posted by Mingle
Which spreadsheet are you referring to? If you're using online nMonte runner, CHG is definitely there.Quote:
Originally Posted by Toppo900
Yes, low distances are good when using ancient samples.Quote:
Originally Posted by Toppo900
Wait..um, if the distance say is greater than 1.5 is good fit when using ancient samples but bad if we're using modern samples, right? I just want to get the confusion out. This is the calulcation I've done for Palestinian Muslims using ancient samples:Quote:
Originally Posted by Mingle
"sample": "Test1:Palestinian",
"fit": 2.4035,
"Levant_PPNB": 50.83,
"Kura-Araxes_ARM_Kalavan": 31.67,
"IRN_Ganj_Dareh_N": 10.83,
"TZA_Pemba_600BP": 5,
"MAR_Iberomaurusian": 1.67,
"Anatolia_Barcin_N": 0,
To you, this is great fit, right?
It's not a great fit, it's okay.Quote:
Originally Posted by Toppo900
For ancient samples, the lower the better. But if you get high distances using ancients, I wouldn't get too worried since genetic drift causes older samples to naturally be more distant.
For modern samples, try to keep it at around 2.
Found it. It's in Georgia category.Quote:
Originally Posted by NPKTO
"sample": "Custom:Palestinian",
"fit": 3.7165,
"Levant_PPNB": 55.83,
"IRN_Wezmeh_N": 22.5,
"Anatolia_Barcin_N": 11.67,
"GEO_CHG": 10,
I think it needed more simple in the mix.
0.01
"sample": "Test1:Palestinian",
"fit": 3.666,
"Levant_PPNB": 54.17,
"IRN_Ganj_Dareh_N": 19.17,
"GEO_CHG": 12.5,
"Anatolia_Barcin_N": 8.33,
"MAR_Iberomaurusian": 5.83,
0
"sample": "Test1:Palestinian",
"fit": 3.604,
"Levant_PPNB": 52.5,
"IRN_Ganj_Dareh_N": 22.5,
"GEO_CHG": 10,
"Anatolia_Barcin_N": 9.17,
"MAR_Iberomaurusian": 5.83,
How is it now?
Which region of Palestine is the OP from by the way?
But..isn't 2 a high distance for the fit to be good using modern samples?Quote:
Originally Posted by Mingle
"sample": "Test1:Palestinian",
"fit": 1.3939,
"Lebanese_Christian": 65.83,
"Egyptian": 20.83,
"BedouinB": 10.83,
"Dinka": 2.5,
I don't get it. Does this look like a bad fit to you? By your statements, this is a pretty good fit for Palestinians when using only ancient samples:
"sample": "Test1:Palestinian",
"fit": 1.0421,
"Levant_Canaanite_MBA": 53.33,
"Levant_JOR_EBA": 22.5,
"TKM_IA": 13.33,
"Canary_Islands_Guanche": 6.67,
"TZA_Pemba_600BP": 4.17,
"GEO_CHG": 0,
In other words, the lesser the distance the greater the accuracy when using the ancient samples for moderns, and the greater the distance the greater the accuracy when using the modern samples for moderns?
The West Bank.Quote:
Originally Posted by Mingle
This man is a typical Palestinian just less Arab shifted than the average.
God, you guys are making things confusing for me.
It looks like there's Steppe as well. Using Yamnaya lowered the fit down. CHG is not required in this model. The other model you posted is good I think.Quote:
Originally Posted by Toppo900
"fit": 3.0772,
"Levant_PPNB": 54.17,
"IRN_Ganj_Dareh_N": 25,
"Anatolia_Barcin_N": 9.17,
"Yamnaya_RUS_Samara": 7.5,
"MAR_Iberomaurusian": 4.17,
I used more recent ancient samples for the Palestinian Muslims:Quote:
Originally Posted by NPKTO
"sample": "Test1alestinian",
"fit": 1.0421,
"Levant_Canaanite_MBA": 53.33,
"Levant_JOR_EBA": 22.5,
"TKM_IA": 13.33,
"Canary_Islands_Guanche": 6.67,
"TZA_Pemba_600BP": 4.17,
"GEO_CHG": 0,
According to Mingle, this is very accurate.
You just have to be very careful not to have multiple closely related components.Quote:
Originally Posted by Toppo900
They're aren't closely related in the results, so it's good.Quote:
Originally Posted by NPKTO