100% probability Scythians ancestral to Kurds & some Iranians according to big study

[axlredneck]

What's your point. They're not allowed to post any references that talk about Mt-DNA

Check the references in the supplementary information.

[Zoro]

I'm from Ukraine

Target: Cumansky
Distance: 3.7661% / 0.03766134
29.2 NOR_N_HG
26.4 Saka_Tian_Shan
24.6 MKD_N
12.2 Kurdish
4.6 Gambian
2.0 Fulani
1.0 BedouinB

Yeah, Ukrainians and other E Europeans have quite a bit of W Scythian ancestry. I'm not saying this based on what looks like a G25 output because those types of programs don't parse the various Scythians well and they are not even informative to the direction of geneflow. So for example if Saka Tian Shan hypothetically had 26% ancestry related to Ukrainians it would still show what you are showing.

BTW, if I was Ukrainian I would start out with logical sources such as the Ukrainian or other E European or Caucasus Scythian samples along with some MLBA steppe along with maybe E European Chl and maybe SW Asian sources

[Zoro]

If you're stubborn and dont wanna accept the truth, well, don't! I gave you the reference from the same Table 19, and from the same Supplementary Information, and no where else.

If they wanted to compare Scythian Mt-DNA to the 500 contemporary samples they would list the Mt-DNA for each sample or show the Mt-DNA breakdown for each contemporary population.

Do yourself a favor and be guided by the descriptions for each figure and don't blanket assume every damn chart and figure in the paper is about Mt-DNA. In fact the opposite is true. If you carefully read each table or chart description you'll see that most of them are about autosomes. No one ever does dstats or f4s or f3s on Mt-DNA.

Enough time wasted on 1st grade stuff

[Cumansky]

Yeah, Ukrainians and other E Europeans have quite a bit of W Scythian ancestry. I'm not saying this based on what looks like a G25 output because those types of programs don't parse the various Scythians well and they are not even informative to the direction of geneflow. So for example if Saka Tian Shan hypothetically had 26% ancestry related to Ukrainians it would still show what you are showing.

BTW, if I was Ukrainian I would start out with logical sources such as the Ukrainian or other E European or Caucasus Scythian samples along with some MLBA steppe along with maybe E European Chl and maybe SW Asian sources

I ran some model just now, when I remove Saka Tian Shan from spreadsheet and run it isolated was picking a low distance around 1, majority of genome was Kurdish, a smaller part was picking up Mongolian ancient samples, and a trace Baltic North European

For me Saka Tian Shan in this basic run pick up Eurasian ancestry mostly Caucasoid some trace Mongoloid but also something that went into Western Europeans probably Sintasha related if I had to bet

But I have low West Euro ancestry we went to war with Nazi, so whatever is in the Western Euros and Kurds probably also survive from Cossacks

[axlredneck]

If they wanted to compare Scythian Mt-DNA to the 500 contemporary samples they would list the Mt-DNA for each sample or show the Mt-DNA breakdown for each contemporary population.

Do yourself a favor and be guided by the descriptions for each figure and don't blanket assume every damn chart and figure in the paper is about Mt-DNA. In fact the opposite is true. If you carefully read each table or chart description you'll see that most of them are about autosomes. No one ever does dstats or f4s or f3s on Mt-DNA.

Enough time wasted on 1st grade stuff

S. Figure 10 and 11 direct you to S. Table 19 for reference populations, and the latter directs to the references for information relating to each and every ethnicity mentioned in the table. Each reference is about MT-DNA and not autosomal related! Plus, they wouldn't have to compare each and every ethnicity separately to show that this ethnicity has this percentage of mt-DNA. They've made charts and that serves the purpose!

[Kaspias]

Using the stuff like G25 is better while determining potential ancestry, it is open-ended, unlike dstats and f. These are good while seeing geneflow as you stated but before check geneflow, you have to be sure that you have this admixture. However, I can't see such input at the beginning. Simple, if I would model myself here with Haji-Firuz Chl, W-Scythian, I would be around 80% Scythian and it would really show my geneflow. But this does not necessarily reflect reality.

I think optimist, and believe that these are all accurate, then it also should show up in G25.



Avg comparison with other Iranians:

[Zoro]

S. Figure 10 and 11 direct you to S. Table 19 for reference populations, and the latter directs to the references for information relating to each and every ethnicity mentioned in the table. Each reference is about MT-DNA and not autosomal related! Plus, they wouldn't have to compare each and every ethnicity separately to show that this ethnicity has this percentage of mt-DNA. They've made charts and that serves the purpose!

Sure boss the whole fucking paper is nothing but mt-dna and sure they don’t have to give the mt-dna distribution or the mt haplogroup of the 500 modern samples they’re comparing the scythians to . We can all just guess what their mt-dna distribution or haplogroup is. I mean that’s how scientific papers are written....

All the fucking Admixture plots are mt-dna based
All the f3 graphs are mt-dna based
All the f4 graphs are mt-dna based
All the qpwave and qpAdm outputs are mt-dna based. Who cares if those programs are built for autosomal inputs. What do Lazaridis and Reich know about these programs. Their institutions just happen to build them

And kurd mt-dna is all Scythian. Happy?

Now get lost and stop cluttering this thread with useless rubbish

[axlredneck]

Sure boss the whole fucking paper is nothing but mt-dna and sure they don’t have to give the mt-dna distribution or the mt haplogroup of the 500 modern samples they’re comparing the scythians to . We can all just guess what their mt-dna distribution or haplogroup is. I mean that’s how scientific papers are written....

All the fucking Admixture plots are mt-dna based
All the f3 graphs are mt-dna based
All the f4 graphs are mt-dna based
All the qpwave and qpAdm outputs are mt-dna based. Who cares if those programs are built for autosomal inputs. What do Lazaridis and Reich know about these programs. Their institutions just happen to build them

And kurd mt-dna is all Scythian. Happy?

Now get lost and stop cluttering this thread with useless rubbish

You have the comprehension skills of a fucking amoeba!
You get lost instead.

[Zoro]

Using the stuff like G25 is better while determining potential ancestry, it is open-ended, unlike dstats and f. These are good while seeing geneflow as you stated but before check geneflow, you have to be sure that you have this admixture. However, I can't see such input at the beginning. Simple, if I would model myself here with Haji-Firuz Chl, W-Scythian, I would be around 80% Scythian and it would really show my geneflow. But this does not necessarily reflect reality.

I think optimist, and believe that these are all accurate, then it also should show up in G25.



Avg comparison with other Iranians:

A couple of points real quick:

1- It’s never a good idea to use a program such as ADMIXTURE or PCA to calculate ancestry where the program user throws 1000s of samples in the run because the 1000s of samples have an UNKNOWN push or pull on the target you’re trying to model. That’s why if Davidski had run the PCA with only the samples which make sense for each specific target or the ADMIXTURE program is only run using logical sources for your specific target the G25 coordinates or ADMIXTURE output would be totally different and make more sense. But these people will not run them this way because it’s not practical for them since they would have to do 100s of PCA or ADMIXTURE runs one for each specific population and tell the users if you belong to population A you can only use calculator A or G25 coordinates obtained using only sources logical for population A

You’ll notice that’s exactly what they do in scientific papers when modelling someone using dstats f3s or qpAdm theyonly use a couple of sources and outgroups in each model

2- G-25 or Admixture doesn’t tell you standard error or p-value or model likelihood for EACH specific model. Distance is not an acceptable measure

3- The formal models f3s f4s qpAdm and so on including the paper in the OP contradict what you just posted. Sorry,but I’ll go with them over G25 or ADMIXTURE for the reasons mentioned and more any day.

[Zoro]

I would be around 80% Scythian and it would really show my geneflow. But this does not necessarily reflect reality.

Not if you have someone who knows what they’re doing model you with qpAdm using the logical sources for your group.maybe anatolia chl, bulgaria chl, haji firuz, and various scythians and Sarmatian s