Attachment 146512Quote:
Originally Posted by Gannicus
Very close again
Printable View
Attachment 146512Quote:
Originally Posted by Gannicus
Very close again
Have you tried it with the references I used in post #399? There’s probably a good chance it will be the same for you also.Quote:
Originally Posted by Birchy
Yh exact same referencesQuote:
Originally Posted by Gannicus
What do you get when you swap out Yamnaya for Latvia_LN_CordedWare.AG ?Quote:
Originally Posted by Birchy
Me:
Attachment 146516
References:
Attachment 146517
Attachment 146518
Pretty sure these are the same references.
Attachment 146524Quote:
Originally Posted by Gannicus
You all can try this and see what you get.
Me:
Attachment 146525
Attachment 146526
References by AndreiDNA, I just added Dzudzuana:
Attachment 146527
Edit: I tried this without ANF and used Levant N instead with the same references.
Attachment 146528
One thing I like about this run is that it brings my WHG admix closer to where I think it should be.
Finally got around to doing this lolQuote:
Originally Posted by drb234
Attachment 146556
Also tried Denmark IA
Attachment 146558
Try this small reference list:
Attachment 146684
What I get with Yamnaya Steppe, Barcin ANF, and Loschbour:
Attachment 146685
Same model but with different rights to reduce SEs, interestingly I seem more Gaulic shifted here
Attachment 146721
Attachment 146722
Attachment 146723
Attachment 146724
Attachment 146725
Attachment 146726
Model of Anglo Canadians
AndreiDNA put out this video:
https://www.youtube.com/watch?v=K8gSZ71h5_M&t=207s
So, I asked him to model me with the populations used in his Admixtools 2 model at 3:28.
It may not be the same rights from the video
Attachment 146756
Rights used:
Mbuti.DG,Sweden_LN.SG,England_IA.AG,Lithuania_EMN_ Narva.AG,Croatia_Popova_MN.SG,Norway_IA.SG,Scotlan d_LBA.AG,Estonia_EarlyViking.SG,Czechia_MBA_Tumulu s.AG,England_N.AG,Sweden_BattleAxe.SG,Poland_Maslo mecz_Wielbark_IA.SG,France_LaClape_LN_EMBA.SG,Swed en_IA_2.SG,Croatia_Jagodnjak_MBA.SG,Slovakia_N_LBK .AG,Scotland_BlackIsle_IA.SG,Czechia_EBA_CordedWar e.AG,Sweden_south_Pre_Viking.SG
I decided to try this with the same rights in qpAdm on IllustrativeDNA. The one difference is I'm using my DNA file from MyHeritage instead of my Merged DNA file:
Attachment 146757
I had Andrei then use Norway IA instead of Denmark IA:
Attachment 146758
rights:
Mbuti.DG,Sweden_LN.SG,England_IA.AG,Lithuania_EMN_ Narva.AG,Croatia_Popova_MN.SG,Denmark_IA.SG,Scotla nd_LBA.AG,Estonia_EarlyViking.SG,Czechia_MBA_Tumul us.AG,England_N.AG,Sweden_BattleAxe.SG,Poland_Masl omecz_Wielbark_IA.SG,France_LaClape_LN_EMBA.SG,Swe den_IA_2.SG,Croatia_Jagodnjak_MBA.SG,Slovakia_N_LB K.AG,Scotland_BlackIsle_IA.SG,Czechia_EBA_CordedWa re.AG,Sweden_south_Pre_Viking.SG
And here is what it looks like in IllustrativeDNA qpAdm with just MyHeritage file:
Attachment 146759
With just France Aube IA and Norway IA
Attachment 146760
Here are my genetic distances measured using the f2 statistic.
Attachment 146778
Attachment 146779
Attachment 146780
Attachment 146781
CEU samples are Anglo-Americans from Utah I believe.
I forgot to mention AndreiDNA did this for me.
Here's how FST looks in comparison to my f2:
Attachment 146783
OpenAI interpretation:
This is actually a very clean set of distances.
Let’s look at them carefully.
Your FST Distances (Ordered)
Population FST Distance
English.HO 0.0002
CEU.DG 0.0018
French.HO 0.0024
Norwegian.HO 0.0034
Spanish.HO 0.0034
Czech.HO 0.0039
1️⃣ The English Result Is Extremely Close
0.0002 is extremely tight.
For context:
Within-population variation often falls around ~0.000–0.001
Distances under 0.002 are very small
Under 0.001 is extremely close
0.0002 means you are essentially sitting right on top of the English reference cluster.
That’s not “loosely British.”
That’s strongly aligned.
2️⃣ CEU at 0.0018
CEU (Utah residents with Northern/Western European ancestry) is:
A diaspora sample
Historically mixed between British, German, Scandinavian, etc.
The fact you are closer to modern English than CEU reinforces something:
You are slightly more tightly aligned to the British Isles cluster than the average CEU sample.
3️⃣ French at 0.0024
This also makes total sense.
You have documented French/Belgian ancestry.
Northern French genetically overlap heavily with southern England.
0.0024 is still very close.
4️⃣ Norwegian / Spanish / Czech
As distance increases:
0.0034 Norwegian → North Sea shift
0.0034 Spanish → Atlantic shift
0.0039 Czech → Central European shift
You are still clearly in Northwest Europe, not Central or Southern.
What This Means Structurally
Your genome is sitting in the Northwest European core cluster, slightly biased toward:
English > CEU > French
That pattern is exactly what we would expect from:
~55–60% British Isles
Remaining ancestry from NW Continental Europe
No Southern/Eastern European pull
Because culturally, Americans often think:
“I’m a mix of everything.”
But genetically, your mix is very regionally tight.
Instead of being evenly split across distant European regions, you are consolidated into one specific cluster.
The numbers are confirming that.
Important Perspective
Your English distance (0.0002) vs CEU (0.0018):
The difference is small in absolute terms, but directionally meaningful.
It tells us:
You are not generically “white American.”
You are clustering more tightly with modern English than with a mixed Anglo-American reference.
That aligns very well with your reconstructed great-grandparent breakdown.
The Big Takeaway
Your intuition wasn’t wrong.
Your paper trail suggested a strong British component.
Your FST confirms:
This is population structure showing a consolidation effect.
Compare to IllustrativeDNA's PCA system:
Attachment 146784
Attachment 146785
Modeled myself as 100% English and 100% Scottish with a somewhat small list of 10 references.
Attachment 146896
Attachment 146897
Compared to French:
Attachment 146898
References:
Attachment 146899
remove
Attachment 146952Quote:
Originally Posted by Xavier727
For the record I also plot quite closely on the Celtic side on G25 too, so I'm not sure where the elevated Germanic comes from in Qpadm.
My FST genetic distances with my merged DNA file. Courtesy of AndreiDNA.
Attachment 146982
Attachment 147003
Attachment 147004
Attachment 147005
Some of the strongest ancient models of myself I’ve managed after experimenting quite a bit with the references :)
This FST has some different populations in it from the previous one.
Attachment 147002
That Scottish sample is weird with this analysis method.
Have you done FST on yourself yet?Quote:
Originally Posted by Birchy
What I get with those sources you used:
Attachment 147006
This one is with Drb's references:
Attachment 147007
I will try this laterQuote:
Originally Posted by Gannicus
https://i.postimg.cc/bw3XLhy7/earlyanglosaxons.png
model of early 'unmixed' anglo saxons in england
What were the reference pops you used on this ?Quote:
Originally Posted by drb234
right = c("Mbuti.DG", "EarlyCW_LN1_poz81.SG","Netherlands_BellBeaker.AG" , "Croatia_Jagodnjak_MBA.SG", "Czechia_EBA_Unetice.AG", "France_MN.AG", "Denmark_LN_BA.SG", "Denmark_SouthScandinavia_EBA.SG", "Czechia_MBA_Tumulus.AG", "Austria_N_LBK.AG", "Netherlands_LNB_BellBeaker.AG")Quote:
Originally Posted by Xavier727
Thanks for thatQuote:
Originally Posted by drb234
Models of modern north English groups. Would translate to roughly 55-60% early Anglo-Saxon
https://i.postimg.cc/rsRgbz0T/English-Yorkshire.png
https://i.postimg.cc/1XVMT4ny/englisheastmids.png
What is France_IA_Hallstatt.SG ? Is it the HautsdefranceIA2 samples? Same with the Denmark_Jutland_IA.SG. I'm assuming IllustrativeDNA has those, just labeled differently.Quote:
Originally Posted by drb234
illustrative doesnt, they're from a larger dataset that i useQuote:
Originally Posted by Gannicus
That's disappointing. You'd think they would regularly update the dataset on there.Quote:
Originally Posted by drb234
I've been experimenting with those rights.Quote:
Originally Posted by drb234
Attachment 147272
Attachment 147273
Standard errors aren't that great, but I wonder if the p value would cause some to ignore it.
And then I tried just using 2 sources:
Attachment 147274
Attachment 147275
Slightly better with just 2 sources. If one shoots for Z scores 3 or close to it.
Still experimenting. This is what I have for now.
low standard errors > p value (unless p val is below 0.1/0.05ish)Quote:
Originally Posted by Gannicus
i dont think that set of rights would work well for ancestry/23andme files. they're pretty low coverage so they're difficult to model accurately as a result
delete
modeled Beethoven (who was half flemish / half rhenish german)
https://i.postimg.cc/65L9RgtZ/image.png
On Illustrative for qpAdm I'm using just my DNA file from MyHeritage.com.Quote:
Originally Posted by drb234
I had to add a few more references to get those SEs down a bit more.
Attachment 147287
Attachment 147288
Attachment 147289
Attachment 147313
Austria IA La Tene + Norway IA
In the spirit of St Patrick's Day, I modeled myself as 100% medieval Irish.
Attachment 147317
References:
Attachment 147318
Attachment 147319
Attachment 147321
^ Austria IA LaTene + Norway IA to compare to Birchy
I used Drb's rights from a couple posts ago.
https://www.youtube.com/watch?v=FCnGD6xv5ikQuote:
Originally Posted by Gannicus
Attachment 147353
Denmark IA + Scotland IA (with DRB’s refs)
Modeled myself as 100% VK202. The dataset on IllustrativeDNA has him labeled as: Pict (Orkney).
Attachment 147371
Rights:
Attachment 147372
Attachment 147373
Attachment 147374
I also match him on 23&Me's Historical matches
Attachment 147375
Attachment 147376
I don't know if he's 100% Pictish ancestry or mostly with some Scandinavian. From what I was able to look up he is majority Pictish admixture.
France Yonne IA + Norway IA with Drb's newest rights:
Attachment 147377
Z scores would be over 3 with this model.
Modified Drb rights model:
Attachment 147378
Attachment 147379
Attachment 147380