The relationship of R1a, R1b and R2 in populations: ancient times up to the 1400s

[TigerMW]

This thread is probably long overdue. These guys are R's for a reason. They all have M207+ and no one else does.

The research titled "New binary polymorphisms reshape and increase resolution of the human Y chromosomal haplogroup tree" (2008), Karafet et al. estimated the R TMRCA as 26.8k ybp and the R1 TMRCA as 18.5k ybp. This is probably a better update on the aging than from Wells earlier study.

In "The Eurasian Heartland: A continental perspective on Y-chromosome diversity" (2001), Wells et al., he has Central Asia as being the origin for the R haplogroup and R, R1a, R1b and R2 if I remember correctly.

Here is my understanding of the very high level tree. Let me know if there are errors with it. I don't know above and beside R1b that well.




You've seen the chart below before, but here is the R1b-ht35 administrator and citizen-scientist Vince Vizachero's geographical depiction of the early branching of R.




I rarely get a chance to view and contemplate this whole chart, but here is the high level view of the ISOGG version of haplogroup R's tree:
http://isogg.org/tree/ISOGG_HapgrpR.html

[EDIT on 07/22/2013: I just want to be clear that even though I showed the very early branching of R as proposed by Vince Vizachero I just included that as background for the topic. I'm not intending to restrict this topic to just the 20k to 15k ybp. In fact I think some of the most interesting relationships might be of R1a elements and R1b elements in populations in SW Asia and Central Asia and/or the Balkans during the metal ages. I apologize for my lack of clarity in describing "ancient".]

[alan]

well dating would be crucial. Intepretation options are totally depended on date when you include climate, the situation with the large seas, subsistance economy etc. I certainly doubt M73 went west into Anatolia and entered Europe from there. For a start there is basically none in the Balkans. Think that is based on 1% of Turks, a people well known to also have m73 on their ancestral path west. Far easier IMO to place it nearer its present distribution and intermediate between that and M269's distribution. The north Caucasus or somewhere in the steppe nearby seems more likely to me. P297 is ancestor to both and I would again place it somewhere similar or northern Iran which would explain its nonthingness in the 4000 years between P297 and M73/M269. When it comes to upstream of P297 into the P25* period it gets harder to even guess. However, this is the pre-10000BC period a paper I recently posted did state that the central, south and eastern Caucasus and Iran have a striking lack of upper palaeolithic settlement before 10000BC. So that zone is not a likely area for a very old clade like P25 to have been initially located and more likely an area it might have spilled into several thousand years into its existence. That said it may have spilled into Iran at some point because a recent paper on Iran showed an unusually high amount of very upstream (P297 negative) R1b in the north of Iran. But on present archaeological evidence its unlikely to have first arisen there and is probably a later movement there despite the upstream form.

[Silesian]

This wiki entry should be corrected, as per the above cited study, is similar to Talysh.
"Genetically, the Gilaks display a high frequency of Y-DNA haplogroups R1a1a, J2a, J1, and G2a3b.[4]"
"Caspian languages are a branch of Northwestern Iranian languages spoken in northern Iran, south of the Caspian Sea."


Regards Brahui people, R1a
https://en.wikipedia.org/wiki/Brahui_people

"Another theory is that they migrated to Baluchistan from inner India during the early Muslim period of the 13th or 14th centuries.[5] A third theory says the Brahui migrated to Balochistan from South India after 1000 AD. The absence of any older Iranian (Avestan) influence in Brahui supports this hypothesis. The main Iranian contributor to Brahui vocabulary is a northwestern Iranian language,"

S_Tlsh 18N 44% Indo-Iranian (IE) Talysh Roewer et al.,
Gilaki 43N 23% Indo-Iranian (IE) Roewer et al.
Mazan 46N 15% Indo-Iranian (IE)Mazandarani Roewer et al.
N_Tlsh 43N 19% Indo-Iranian (IE) Talysh Roewer et al.

Gilaki/Talysh/Lurs.

[DMXX]

In an ideal world, we'd have reliable node calculations and enough ancient DNA to definitively piece together how Y-DNA R disseminated itself across Eurasia. Unfortunately, we're not living in an ideal world. The best we can make do with, in my view, is the analysis and interpretation of parahaplogroups to give us a rough pointer as to where the formative steps for each successive downstream marker took place.

R-M207 was sporadically reported in scientific literature as well as among some FTDNA projects (my own Iranian DNA Project once had an R*-M207 individual originally from Badakhshan in Tajikistan-Afghanistan). Following the publication of R2*-M479, vineviz (who also manages the parahaplogroups project I believe) seemed confident that most (if not all) of the previously reported R*-M207 found would turn out to be M479+.

I have already posted data from Grugni et al. showing important R parahaplogroups (R1*-M173, R1a1*-SRY1532, R1b*-M343, R1b1a2*-M269, R1b1a2a*-L23) were found scattered across the Iranian plateau. Underhill et al. also discovered R1a*-M420 across the entire "Near-East" (Iran, Oman etc.) and it was completely absent in India. An early study from the Caucasus (an early Dr. Nasidze paper?) found R1a1*-SRY1532(xR1a1a-M17) in Armenia of all places, where R1a-derived subclades barely reach 3% in total.

All of this implies that...
- Central Asia was the location from which the major division formed within Y-DNA R into two clades, R1-M173 and R2-M479.
- R1 then seemed to have differentiated somewhere a bit further west, likely Iran.
- The picture becomes murkier at this point, with the presence of R1a1* and R1b1* in different countries and not just Iran.

I won't comment on R1a because my understanding of its' current structure is outdated. I still recall the unusual R1a1ax subclades found in South-Central Asia and don't know how they stack up now.

I therefore envision the dissemination of Y-DNA R took a path very similar to vineviz's chart, although I'd put "R1a" around Iran, "R" and "R2" close together in Central Asia and everything else pushed perhaps a bit more westerly.

For the record, I once considered India to have a significant role concerning developments within Y-DNA R. I no longer hold that opinion following the production of data showing Indian R subclades to be derived from elsewhere; all Indian R1a to date is Z93+, the picture concerning R2a hints towards an early L295+ wave followed by L295- intrusions further north, any R1b I've seen around the subcontinent in studies share the same haplotypes such as Nepali R1b etc.

[alan]

This wiki entry should be corrected, as per the above cited study.
"Genetically, the Gilaks display a high frequency of Y-DNA haplogroups R1a1a, J2a, J1, and G2a3b.[4]"
"Caspian languages are a branch of Northwestern Iranian languages spoken in northern Iran, south of the Caspian Sea."


Regards Brahui people, R1a
https://en.wikipedia.org/wiki/Brahui_people

"Another theory is that they migrated to Baluchistan from inner India during the early Muslim period of the 13th or 14th centuries.[5] A third theory says the Brahui migrated to Balochistan from South India after 1000 AD. The absence of any older Iranian (Avestan) influence in Brahui supports this hypothesis. The main Iranian contributor to Brahui vocabulary is a northwestern Iranian language,"

I take it that they have ruled out V88 - yes they did

[alan]

The possibly role of northern Iran in R1b history first really came into my view in after the paper suggesting a link with Maykop. Having read into it I see the flow could have gone both ways and there are Makop kurgans in NW Iran. Also having read into Iran a bit I see that the northern plateau unlike zagros was only very late involved in the farming world.That makes the location a bit more R1b compatible IMO. Howevere I then read that northern Iran was along with the south and east Caucausus apparently not settled by modern humans until after 10000BC. That would rule it out as the homeland for R1b. On balance I think its easier to see the whole r1b sequence as occurring somewhere further north than Iran with Iran being the main southern 'out' for stray P25 into north Iran after 10000BC and further into the middle east after farming reached the iran area and trade networks were under way c. 4000bc.

[TigerMW]

well dating would be crucial. Intepretation options are totally depended on date when you include climate, the situation with the large seas, subsistance economy etc. I certainly doubt M73 went west into Anatolia and entered Europe from there. For a start there is basically none in the Balkans. Think that is based on 1% of Turks, a people well known to also have m73 on their ancestral path west. Far easier IMO to place it nearer its present distribution and intermediate between that and M269's distribution. The north Caucasus or somewhere in the steppe nearby seems more likely to me. P297 is ancestor to both and I would again place it somewhere similar or northern Iran which would explain its nonthingness in the 4000 years between P297 and M73/M269. When it comes to upstream of P297 into the P25* period it gets harder to even guess. However, this is the pre-10000BC period a paper I recently posted did state that the central, south and eastern Caucasus and Iran have a striking lack of upper palaeolithic settlement before 10000BC. So that zone is not a likely area for a very old clade like P25 to have been initially located and more likely an area it might have spilled into several thousand years into its existence. That said it may have spilled into Iran at some point because a recent paper on Iran showed an unusually high amount of very upstream (P297 negative) R1b in the north of Iran. But on present archaeological evidence its unlikely to have first arisen there and is probably a later movement there despite the upstream form.

I would think that there would be some R1a in at least some of the communities that R1b-M73 was in when it expanded and moved. Is there much evidence for that? If so what kinds of R1a?

[newtoboard]

I would think that there would be some R1a in at least some of the communities that R1b-M73 was in when it expanded and moved. Is there much evidence for that? If so what kinds of R1a?

Maybe but these seem to have not been mixed imo. If the R1b-M73 community was in the North Caucasus or near it I some Z283+ (and Z280+) would have have likely moved with it into Central Asia and the Tarim. As far as I know there isn't much Z283+ or Z280+ in the Tarim or Central Asia (some does exist in Uzbekistan/Kazakhstan but that could be Russian admixture since M458+ was also found in Uzbekistan/Kazakhstan. It would also suggest some sort of bottleneck occurring on the area once encompassed by the BMAC). The Tajiks, Turkmens, and Uzbeks all live in a similar area but the Tajiks lack M73+ (while Uzbeks and Turkmen carry it) despite the fact that they were once spread between Afghanistan, Tajikistan, Turkmenistan, Uzbekistan and Kyrgyzstan (countries where the Turkic speakers do carry R1b-M73).

[newtoboard]

The possibly role of northern Iran in R1b history first really came into my view in after the paper suggesting a link with Maykop. Having read into it I see the flow could have gone both ways and there are Makop kurgans in NW Iran. Also having read into Iran a bit I see that the northern plateau unlike zagros was only very late involved in the farming world.That makes the location a bit more R1b compatible IMO. Howevere I then read that northern Iran was along with the south and east Caucausus apparently not settled by modern humans until after 10000BC. That would rule it out as the homeland for R1b. On balance I think its easier to see the whole r1b sequence as occurring somewhere further north than Iran with Iran being the main southern 'out' for stray P25 into north Iran after 10000BC and further into the middle east after farming reached the iran area and trade networks were under way c. 4000bc.

What is your route for more downstream clades making it back into Iran?

[alan]

In an ideal world, we'd have reliable node calculations and enough ancient DNA to definitively piece together how Y-DNA R disseminated itself across Eurasia. Unfortunately, we're not living in an ideal world. The best we can make do with, in my view, is the analysis and interpretation of parahaplogroups to give us a rough pointer as to where the formative steps for each successive downstream marker took place.

R-M207 was sporadically reported in scientific literature as well as among some FTDNA projects (my own Iranian DNA Project once had an R*-M207 individual originally from Badakhshan in Tajikistan-Afghanistan). Following the publication of R2*-M479, vineviz (who also manages the parahaplogroups project I believe) seemed confident that most (if not all) of the previously reported R*-M207 found would turn out to be M479+.

I have already posted data from Grugni et al. showing important R parahaplogroups (R1*-M173, R1a1*-SRY1532, R1b*-M343, R1b1a2*-M269, R1b1a2a*-L23) were found scattered across the Iranian plateau. Underhill et al. also discovered R1a*-M420 across the entire "Near-East" (Iran, Oman etc.) and it was completely absent in India. An early study from the Caucasus (an early Dr. Nasidze paper?) found R1a1*-SRY1532(xR1a1a-M17) in Armenia of all places, where R1a-derived subclades barely reach 3% in total.

All of this implies that...
- Central Asia was the location from which the major division formed within Y-DNA R into two clades, R1-M173 and R2-M479.
- R1 then seemed to have differentiated somewhere a bit further west, likely Iran.
- The picture becomes murkier at this point, with the presence of R1a1* and R1b1* in different countries and not just Iran.

I won't comment on R1a because my understanding of its' current structure is outdated. I still recall the unusual R1a1ax subclades found in South-Central Asia and don't know how they stack up now.

I therefore envision the dissemination of Y-DNA R took a path very similar to vineviz's chart, although I'd put "R1a" around Iran, "R" and "R2" close together in Central Asia and everything else pushed perhaps a bit more westerly.

For the record, I once considered India to have a significant role concerning developments within Y-DNA R. I no longer hold that opinion following the production of data showing Indian R subclades to be derived from elsewhere; all Indian R1a to date is Z93+, the picture concerning R2a hints towards an early L295+ wave followed by L295- intrusions further north, any R1b I've seen around the subcontinent in studies share the same haplotypes such as Nepali R1b etc.

I certainly think in longitude terms this is probably giving us a reasonable idea. However, the problem with the latitude aspect I see is that individuals or small groups who entered the the farming world are more likely to have experienced conditions that allowed preservation of early clades than areas like the steppe where essentially the pre-5000BC era appears to have left little trace in terms of clades. Indeed I only give that early date because of M73 and its distribution. In fact it appears that other than that little is preserved of y lineages or clades pre-dating 3500BC.

That may simply be the nature of the non-farming/very basic farmer-herder-hunter societies there and they may have simply struggled to maintain population for a long period and probably tended at any given point to share relatively recent common ancestors. That might only have altered first c. 4200BC when steppe groups first expanded into new lands in Old Europe and again when horses, the wheel, mobility etc opened up large new areas of pasture c 3500BC onwards. Both of these might have allowing a phase of ydna expansion that survived to today. Those date do roughly coincide with the rise of R1a and R1b lineages of importance after a long period of nothingness. So it is possible that the early clade guys we see in the middle east are simply strays that entered the farming area early. I have recently posted that there was a barrier to north-south and east-west movement around the Black-Caspian sea area in the period after the LGM to around 10000BC too so movements in either direction probably had a long hiatus in this sort of longitude until 10000BC and there does seem to indeed be a lack of upper Palaeolithic settlement on the south side of this divide until around that date. There may have been a period in the upper palaeolithic where, counterintuitive though it seems, the north-west side of this barrier may have been much better settled by hunters. Taking the apparent period of those barriers, it seems to me to make the most likely phases of north-south movement at this sort of longitude to be around the LGM before the barrier and around 10000bc after the barrier passed. Given that these essentially coincide with the LGM and the Younger Dryas a north-south movement would seem far more likely than the reverse. Northern Iran did indeed seem to recieve an appearance of epi-palaeolithic hunters at this period. It is also important to remember that northern Iran and its plateau did not recieve farming until rather late too and there is no evidence of a movement from there northwards in either the Palaeolithic or the Neolithic. So, I think northern Iran is perhaps more likely an area that saw overspill from the north around the Younger Dryas than a source of northern R1b or R1a.

So, I tend to think that r1 was already in what is now the western steppe area by the LGM and got bottled up there by post-LGM water barriers to the east and south after the LGM. Karafet dates it to about 16500BC which is in the right ballpark.