R2 - Ancient DNA Discussion

[Tomasso29]

I want to start a topic that discusses R2 ancient DNA on here. Seeing that there are a number of us on this forum and since they found plenty of ancient samples, I think it's worth having an ongoing thread that discusses this topic. I think putting this here is very fitting, given that those other threads in the other sections always get hijacked by folks fighting over a specific topic over and over again.

Anyways, so based on what we found so far, do you guys see anything that may pin point to a potential origin point? Or perhaps a specific culture that R2 is associated with? Imo, the origin point is still up for debate, but R2 seems to be strongly associated with the Iran_N element, any thoughts?

[aaronbee2010]

I want to start a topic that discusses R2 ancient DNA on here. Seeing that there are a number of us on this forum and since they found plenty of ancient samples, I think it's worth having an ongoing thread that discusses this topic. I think putting this here is very fitting, given that those other threads in the other sections always get hijacked by folks fighting over a specific topic over and over again.

Anyways, so based on what we found so far, do you guys see anything that may pin point to a potential origin point? Or perhaps a specific culture that R2 is associated with? Imo, the origin point is still up for debate, but R2 seems to be strongly associated with the Iran_N element, any thoughts?

Good idea! In that case, it's probably time for me to have another look at the ancient samples from the Narasimhan et al. 2009 study. My experience of bioinformatics has improved a little bit since the last time I took a crack at this. Here's one to start (I thought I would start with this sample as it has the closest match to my own Y-DNA of any ancient DNA sample discovered so far).

NOTE: The sample below was aligned from a hg19 BAM to a FASTQ (FASTQ's for I7722 weren't available from their ENA accession) then to a hg38 BAM ("no-alt analysis set" reference) with BWA on usegalaxy.eu.

I7722 (Saidu_Sharif_H_outlier):

YFull Haplogroup: R2-Z29271* (xR2-Y1383)

Positive SNPs - YFull Haplogroup - No. of Derived Reads - No. of Ancestral Reads:

SK2142 - R2-Y1378 - 2 - 0
Z29195 - R2-Y1378 - 1 - 0
Z29275 - R2-Y1378 - 1 - 0
Z29276 - R2-Y1378 - 2 - 0
Z29280 - R2-Y1378 - 7 - 0
Y1379 - R2-Y1376 - 3 - 0
Z29272 - R2-Y1376 - 1 - 0
Z29264 - R2-Z29271 - 1 - 0
Z29271 - R2-Z29271 - 6 - 2 - Both ancestral reads were from fragments with mapping quality (MAPQ) = 0, so can be ignored whereas none of the derived reads had MAPQ=0.
FT67423 - R2-Z29271* - 2 - 0 - This SNP is currently found only with I7722 and NA20887, who is also R2-Z29271* (xR2-Y1383).

Positive SNPs - YFull Haplogroup - No. of Derived Reads - No. of Ancestral Reads:

BY221205 - R2-Y1383 - 0 - 6
Y1382 - R2-Y1383 - 0 - 1
Z29285 - R2-Y1383 - 0 - 2
Z29287 - R2-Y1383 - 0 - 7
Z29289 - R2-Y1383 - 0 - 1
Z29291 - R2-Y1383 - 0 - 1
Z29292 - R2-Y1383 - 0 - 1
Z29300 - R2-Y1383 - 2 - 4 - Both derived reads were from fragments with mapping quality (MAPQ) = 0, so can be ignored whereas none of the ancestral reads had MAPQ=0.
Z29301 - R2-Y1383 - 0 - 1
Z29302 - R2-Y1383 - 0 - 1
Z6135 - R2-Z6135 - 1 - 5

EDIT: On another note, it seems like R2a-M124 originated in or near the Zagros region, but R2-M479 itself appears to have originated around Afghanistan, judging from resemblances between Iran_N and Aq Kupruk (North Afghanistan) in terms of artifacts (megaliths being a particularly good example), and similar dates of the R2-M479 TMRCA when compared to the date of the Aq Kupruk culture. Also, R2b-SK2164 is clustered around South-Central Asia, close to Aq Kupruk. This all seems to suggests that R2-M479 originated somewhere between North Afghanistan and Lake Baikal, but probably noticeably closer to the former.

As far as intra-topic arguments ago, I think the R2's on this forum have been exceptionally sensible in this regard

[aaronbee2010]

Positive SNPs - YFull Haplogroup - No. of Derived Reads - No. of Ancestral Reads:

BY221205 - R2-Y1383 - 0 - 6
Y1382 - R2-Y1383 - 0 - 1
Z29285 - R2-Y1383 - 0 - 2
Z29287 - R2-Y1383 - 0 - 7
Z29289 - R2-Y1383 - 0 - 1
Z29291 - R2-Y1383 - 0 - 1
Z29292 - R2-Y1383 - 0 - 1
Z29300 - R2-Y1383 - 2 - 4 - Both derived reads were from fragments with mapping quality (MAPQ) = 0, so can be ignored whereas none of the ancestral reads had MAPQ=0.
Z29301 - R2-Y1383 - 0 - 1
Z29302 - R2-Y1383 - 0 - 1
Z6135 - R2-Z6135 - 1 - 5

The word in red should be "Negative". Apologies. It's too late for me to edit it now.

[jdonofrio]

I want to start a topic that discusses R2 ancient DNA on here. Seeing that there are a number of us on this forum and since they found plenty of ancient samples, I think it's worth having an ongoing thread that discusses this topic. I think putting this here is very fitting, given that those other threads in the other sections always get hijacked by folks fighting over a specific topic over and over again.

Anyways, so based on what we found so far, do you guys see anything that may pin point to a potential origin point? Or perhaps a specific culture that R2 is associated with? Imo, the origin point is still up for debate, but R2 seems to be strongly associated with the Iran_N element, any thoughts?

I agree with Aaron's posting that the Zagros Mountain region of Iran seems like the most likely region for R2-M124. The Sep 2019 paper by Vagheesh Narasimhan "The Formation of Human Populations in South and Central Asia" includes four ancient R2 samples from the Ganj Dareh site. These have also been carbon dated to about 10000 years ago. I am not aware of any earlier R2 samples than these.

Also, David Reich's book "Who We Are and How We Got Here", on pages 149-152, there is a very interesting map and discussion on the ancient Persian farmers migration, showing the Indian and European clines, and provides some date boundaries as well. It shows a north flow (towards Uzbekistan/Tajikistan/etc) and a southern flow into Pakistan and Indian subcontinent. This is consistent with the other ancient R2 samples from the Narasimhan paper (which it should, as the paper was investigated through Dr. Reich's lab at Harvard.)

I have requested getting the oldest R2 sample loaded to Y-Full, and if the loading is successful, to work on getting other samples loaded. Y-Full has stated they are backlogged with many ancient samples, but they will get to it. So it may take some time. It would be really great to get some of these ancient R2 samples loaded.

By the way, there is an small R2 Y-DNA Haplogroup page on Facebook (less than 50 of us), and we were discussing this topic recently. If anyone is interested, feel free to connect and add to the discussion there as well: https://www.facebook.com/groups/R2YDNA/

[aaronbee2010]

I agree with Aaron's posting that the Zagros Mountain region of Iran seems like the most likely region for R2-M124. The Sep 2019 paper by Vagheesh Narasimhan "The Formation of Human Populations in South and Central Asia" includes four ancient R2 samples from the Ganj Dareh site. These have also been carbon dated to about 10000 years ago. I am not aware of any earlier R2 samples than these.

Also, David Reich's book "Who We Are and How We Got Here", on pages 149-152, there is a very interesting map and discussion on the ancient Persian farmers migration, showing the Indian and European clines, and provides some date boundaries as well. It shows a north flow (towards Uzbekistan/Tajikistan/etc) and a southern flow into Pakistan and Indian subcontinent. This is consistent with the other ancient R2 samples from the Narasimhan paper (which it should, as the paper was investigated through Dr. Reich's lab at Harvard.)

I have requested getting the oldest R2 sample loaded to Y-Full, and if the loading is successful, to work on getting other samples loaded. Y-Full has stated they are backlogged with many ancient samples, but they will get to it. So it may take some time. It would be really great to get some of these ancient R2 samples loaded.

By the way, there is an small R2 Y-DNA Haplogroup page on Facebook (less than 50 of us), and we were discussing this topic recently. If anyone is interested, feel free to connect and add to the discussion there as well: https://www.facebook.com/groups/R2YDNA/

Nice to see you here! It's nice to see someone prominent within the R2 community start posting on Anthrogenica again

[jdonofrio]

Nice to see you here! It's nice to see someone prominent within the R2 community start posting on Anthrogenica again

Thanks Aaron. I haven't checked in here as much as I should have, but I am going to try to more often. I would like to help grow the R2 community, and this is one good place where we R2ers meet, along with the FamilyTreeDNA R2 project, and the Facebook R2 Y-DNA page.

There are lots of good research projects going on these days, it helps to be connected to learn about any new developments in the R2 world.

[indusnomad]

Thanks for sharing that information! It makes sense that R2 originated in the Zagros Mountains. Did it originally come from the steppes? Since R is an Ancient North Eurasian lineage? Also the 2019 paper suggested that there wasn't a large scale migration of Iranian farmers from Iran to the Indus Valley and that farming might've started in the Indus indigenously. Doesn't that contradict the migration pattern of R2 subclades from West Asia to Central and South Asia?

[aaronbee2010]

Good idea! In that case, it's probably time for me to have another look at the ancient samples from the Narasimhan et al. 2009 study. My experience of bioinformatics has improved a little bit since the last time I took a crack at this. Here's one to start (I thought I would start with this sample as it has the closest match to my own Y-DNA of any ancient DNA sample discovered so far).

NOTE: The sample below was aligned from a hg19 BAM to a FASTQ (FASTQ's for I7722 weren't available from their ENA accession) then to a hg38 BAM ("no-alt analysis set" reference) with BWA on usegalaxy.eu.

I7722 (Saidu_Sharif_H_outlier):

YFull Haplogroup: R2-Z29271* (xR2-Y1383)

Positive SNPs - YFull Haplogroup - No. of Derived Reads - No. of Ancestral Reads:

SK2142 - R2-Y1378 - 2 - 0
Z29195 - R2-Y1378 - 1 - 0
Z29275 - R2-Y1378 - 1 - 0
Z29276 - R2-Y1378 - 2 - 0
Z29280 - R2-Y1378 - 7 - 0
Y1379 - R2-Y1376 - 3 - 0
Z29272 - R2-Y1376 - 1 - 0
Z29264 - R2-Z29271 - 1 - 0
Z29271 - R2-Z29271 - 6 - 2 - Both ancestral reads were from fragments with mapping quality (MAPQ) = 0, so can be ignored whereas none of the derived reads had MAPQ=0.
FT67423 - R2-Z29271* - 2 - 0 - This SNP is currently found only with I7722 and NA20887, who is also R2-Z29271* (xR2-Y1383).

Negative SNPs - YFull Haplogroup - No. of Derived Reads - No. of Ancestral Reads:

BY221205 - R2-Y1383 - 0 - 6
Y1382 - R2-Y1383 - 0 - 1
Z29285 - R2-Y1383 - 0 - 2
Z29287 - R2-Y1383 - 0 - 7
Z29289 - R2-Y1383 - 0 - 1
Z29291 - R2-Y1383 - 0 - 1
Z29292 - R2-Y1383 - 0 - 1
Z29300 - R2-Y1383 - 2 - 4 - Both derived reads were from fragments with mapping quality (MAPQ) = 0, so can be ignored whereas none of the ancestral reads had MAPQ=0.
Z29301 - R2-Y1383 - 0 - 1
Z29302 - R2-Y1383 - 0 - 1
Z6135 - R2-Z6135 - 1 - 5

EDIT: On another note, it seems like R2a-M124 originated in or near the Zagros region, but R2-M479 itself appears to have originated around Afghanistan, judging from resemblances between Iran_N and Aq Kupruk (North Afghanistan) in terms of artifacts (megaliths being a particularly good example), and similar dates of the R2-M479 TMRCA when compared to the date of the Aq Kupruk culture. Also, R2b-SK2164 is clustered around South-Central Asia, close to Aq Kupruk. This all seems to suggests that R2-M479 originated somewhere between North Afghanistan and Lake Baikal, but probably noticeably closer to the former.

As far as intra-topic arguments ago, I think the R2's on this forum have been exceptionally sensible in this regard

Update: I had another look at I7722 and I discovered that FGC2051 and a novel variant have also been found both in NA20887 and I7722, but no other Z29271+ samples I've evaluated.

FGC2051 has 5 reads in I7722 and all are derived. This SNP is apparently present in multiple subclades, but none coming under F-M89, at least according to YFull, although they also class it as being the same mutation as CTS4293/PF2675 according to the YFull SNP search despite the latter having the ancestral and derived alleles the wrong way around. What I find rather strange is that when looking at YFulls raw data browser, they do acknowledge this distinction.

The novel variant has hg19 Y coordinate 26494310 and hg38 chrY coordinate 24348163 and represents a change from C to T. This novel only has one read though (and the read fragment in question has three mismatches from the reference (including this novel)) so appears to be a low quality read. While it would be hard to consider on its own, but considering that FT67423 and FGC2051 are exclusive to I7722 and NA20887 (at least as far as R2 is concerned) and NA20887 has 9 derived reads and 1 ancestral read at this position (6 derived reads and no ancestral reads if you remove reads with a MAPQ of less than 20), I think the presence of this mutation in I7722 is more likely to be genuine than not, all things considered.

In any case, FT67423 and FGC2051 alone are more than sufficient in regards to forming a new subclade with I7722 and NA20887.

On a different note, I'm planning on getting around to analysing other ancient R2 DNA samples, as well as R2 samples from the HGDP, but I'm very busy with exam revision and my dissertation, so this will have to wait until after then.