PDA

View Full Version : Frequency/heatmap of R1a-Z93?



Dr_McNinja
02-04-2015, 11:49 AM
Which are the accurate ones?

I've seen this one:

https://lh3.googleusercontent.com/-75LFC2DbVTQ/U0pzZpEe2QI/AAAAAAAAAU4/-KnlCxo9SkM/w746-h408-no/Underhill_Z93.png

My only issue with that is that R1a-Z93 seems like it should be way more common in Afghanistan/Pakistan/etc?

Eupedia's map: http://cdn.eupedia.com/images/content/Haplogroup-R1a-Z93-Eurasia.png

Or does this pattern (L657 maybe?) perhaps correlate to the easternly spread of Eurogenes K7/K8's WHG-UHG component in South Asia? (East of Pakistan mostly across Punjab, Haryana, Nepal).

DMXX
02-04-2015, 08:55 PM
The maps have very similar patterns. The only major difference between them is the displaying of sampling locations. Shown in Underhill's, absent in Eupedia's. Knowing where the samples are is crucial in determining whether the gradients between various frequency peaks and troughs are artefactual or not. Underhill's is better in this respect.

The frequency increments are also similar. Both plots essentially show the same thing from 20-60%. A 5% increment all the way from 0-60% would have been more informative for fleshing out regional differences.

AMS212
09-06-2021, 11:57 AM
AFAIK South Asians have any significant WHG or UHG in the first place(Barring Jats and Nepali Bahuns perhaps).

Hrodric
11-11-2021, 12:01 PM
Which are the accurate ones?

I've seen this one:

https://lh3.googleusercontent.com/-75LFC2DbVTQ/U0pzZpEe2QI/AAAAAAAAAU4/-KnlCxo9SkM/w746-h408-no/Underhill_Z93.png

My only issue with that is that R1a-Z93 seems like it should be way more common in Afghanistan/Pakistan/etc?

Eupedia's map: http://cdn.eupedia.com/images/content/Haplogroup-R1a-Z93-Eurasia.png

Or does this pattern (L657 maybe?) perhaps correlate to the easternly spread of Eurogenes K7/K8's WHG-UHG component in South Asia? (East of Pakistan mostly across Punjab, Haryana, Nepal).

I would say that Underhill's map is more accurate

Coldmountains
11-11-2021, 12:20 PM
I would say that Underhill's map is more accurate

The underhill map is quite misleading if not wrong. For example, the peak of R1a in North Afghanistan is probably from Kunduz Pashtuns, who were settled there mainly in the 19th century (to defend the region against Russian influences and to punish them for rebellions in East Afghanistan) R1a-Z93 generally is highest among Central and East Pashtuns (medieval founder effect and can reach up to 80% in some regions and tribes) and much lower among local Pre-Pashtun in North Afghanistan.

Generally it is very difficult to make such heatmaps because depending on the valley/village/tribe R1a frequencies can fluctuate extremely in the same region. For example some Pamiri or Tajik subgroups are 60+% R1a and others just 10-20%. The same for Pashtuns who can have tribes with 80% R1a and tribes with less than 20%.

Hrodric
11-11-2021, 04:27 PM
The underhill map is quite misleading if not wrong. For example, the peak of R1a in North Afghanistan is probably from Kunduz Pashtuns, who were settled there mainly in the 19th century (to defend the region against Russian influences and to punish them for rebellions in East Afghanistan) R1a-Z93 generally is highest among Central and East Pashtuns (medieval founder effect and can reach up to 80% in some regions and tribes) and much lower among local Pre-Pashtun in North Afghanistan.

Generally it is very difficult to make such heatmaps because depending on the valley/village/tribe R1a frequencies can fluctuate extremely in the same region. For example some Pamiri or Tajik subgroups are 60+% R1a and others just 10-20%. The same for Pashtuns who can have tribes with 80% R1a and tribes with less than 20%.

Population genetics studies are prone to sampling issues as obviously, but Underhill's paper is the most extensive until now regarding R1a, that's why I prefer his heatmap than Eupedia's one. Nonetheless, I believe we shouldn't take heatmaps as 100% accurate, cuz nothing in population genetics is 100% accurate

Coldmountains
11-12-2021, 11:20 AM
Population genetics studies are prone to sampling issues as obviously, but Underhill's paper is the most extensive until now regarding R1a, that's why I prefer his heatmap than Eupedia's one. Nonetheless, I believe we shouldn't take heatmaps as 100% accurate, cuz nothing in population genetics is 100% accurate

Most of the underhill paper except for the fact that he sampled new populations was outdated and inaccurate.

altvred
11-12-2021, 02:09 PM
Most of the underhill paper except for the fact that he sampled new populations was outdated and inaccurate.

Is it the same paper where the conclusion was that R1a originated somewhere within the vicinity of Eastern Turkey and Western Iran based on the frequency of a "basal subclade" of R1a? The same basal subclade that since then has been found in an Andronovo sample from Kazakhstan? Frankly, I've always found it amusing how much aDNA (that is the oldest R1a samples being found in Eastern Europe) has contradicted that hypothesis from Underhill 2014.

Hrodric
11-12-2021, 05:57 PM
Is it the same paper where the conclusion was that R1a originated somewhere within the vicinity of Eastern Turkey and Western Iran based on the frequency of a "basal subclade" of R1a? The same basal subclade that since then has been found in an Andronovo sample from Kazakhstan? Frankly, I've always found it amusing how much aDNA (that is the oldest R1a samples being found in Eastern Europe) has contradicted that hypothesis from Underhill 2014.

It is not about old samples but rather about samples that bear more basal mutations. Underhill's paper had over 16,000 samples and a paper on R1a with such a large coverage is yet to come. I'll be honest though, I don't know if a study has ever been published that would refute Underhill's hypothesis.

Coldmountains
11-12-2021, 06:28 PM
It is not about old samples but rather about samples that bear more basal mutations. Underhill's paper had over 16,000 samples and a paper on R1a with such a large coverage is yet to come. I'll be honest though, I don't know if a study has ever been published that would refute Underhill's hypothesis.

Even if he sampled 1 million people this would be less relevant for the question of the origin of R1a than a handful of ancient R1a samples from the Neolithic and earlier. The earliest R1a was found 12000 years ago in North Russia (without any archaeological or genetic link to East Turkey/West Iran) and that is enough to refute this hypothesis, which entirely was based on modern DNA.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7817100/


PES001 Peschanitsa Arkhangelsk, RUS WeRuHG 10785–10626 B.C Male XY U4a1 R1a5 5.03

Hrodric
11-12-2021, 06:40 PM
Even if he sampled 1 million people this would be less relevant for the question of the origin of R1a than a handful of ancient R1a samples from the Neolithic and earlier. The earliest R1a was found 12000 years ago in North Russia (without any archaeological or genetic link to East Turkey/West Iran) and that is enough to refute this hypothesis, which entirely was based on modern DNA.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7817100/

ok, but the mutation found wasn't basal enough... it's not about the age of the sample

altvred
11-12-2021, 06:45 PM
It is not about old samples
??

What do you mean by "it's not about the old samples"? R1a was restricted entirely to Eastern Europe until the Middle-to-Late Bronze Age when it expanded into the Central Asian Steppe. According to you, that's somehow irrelevant because of the number of tested individuals in Underhill 2014?


It is not about old samples but rather about samples that bear more basal mutations. Underhill's paper had over 16,000 samples and a paper on R1a with such a large coverage is yet to come. I'll be honest though, I don't know if a study has ever been published that would refute Underhill's hypothesis.

On what criterion are mutations appraised as basal or non-basal? Also, it doesn't matter if there were 100,000 modern samples in this study. The so-called basal subclade has been found an individual dated to 1500 BC in Kazakhstan, thousands of km away from its modern distribution and Underhill's proposed R1a homeland...

https://www.yfull.com/live/tree/R-YP4141/


I4773 I4773 Kazakhstan_MLBA_Aktogai 3508 Kazakhstan 46.96586667 80.03473333 5.029


https://goo.gl/maps/Kt8ay3DfpWDfgpTL9

Coldmountains
11-12-2021, 06:46 PM
ok, but the mutation found wasn't basal enough... it's not about the age of the sample

Almost all R1a today is the result of a massive and rapid expansion of CWC and CWC-derived groups (Steppe_MLBA) during the Bronze Age. These migrations not just spread R1a-M417 and R1a-Z93 to various regions of Eurasia but also basal R1a clades which were present in CWC and Andronovo. The basal R1a in West Iran looks like the result of an expansion of Iron Age Iranics from Central Asia and before the arrival of Indo-Iranians no R1a was found in Iran or Turkey yet.

Hrodric
11-12-2021, 07:02 PM
??

What do you mean by "it's not about the old samples"? R1a was restricted entirely to Eastern Europe until the Middle-to-Late Bronze Age when it expanded into the Central Asian Steppe. According to you, that's somehow irrelevant because of the number of tested individuals in Underhill 2014?



On what criterion are mutations appraised as basal or non-basal? Also, it doesn't matter if there were 100,000 modern samples in this study. The so-called basal subclade has been found an individual dated to 1500 BC in Kazakhstan, thousands of km away from its modern distribution and Underhill's proposed R1a homeland...

https://www.yfull.com/live/tree/R-YP4141/


I4773 I4773 Kazakhstan_MLBA_Aktogai 3508 Kazakhstan 46.96586667 80.03473333 5.029


https://goo.gl/maps/Kt8ay3DfpWDfgpTL9

I won´t discuss anymore... believe what you want... have a nice day