I think you just want to fool yourself, just one ancient skeleton in Afghanistan has been analyzed and this haplogroup has been found (it is possible that it is found in several other ancient skeletons in this land too), a subclade of this haplogroup has been also found in the north of Pakistan, another important point is that this haplogroup still exists among the people of this region. This thing that what you or another person in this forum think about this genetic evidence, really doesn't matter, if you believe it is wrong, you should show me valid sources which fundamentally reject this DNA analysis.
You can think what you want about me if you wish, but just so we're clear, they haven't published that paper due to dating issues, bad calls and other troubles. It's still a pre-print, and it was supposed to be finally published not too long ago, yet here we are still the pre-print and no word from them. By the way one of the authors himself said they were using a new Y-calls software that has some serious issues, not to mention that DNA contamination and damaged DNA issues, he answers questions on his Twitter page. Do you know what low coverage means? Do you know the implications of a + SNP call when the upstream SNPs have a - call? Do you know what that means? It means that the + call is most definitely incorrect. Do you really think it is possible for someone to be R-P311 but negative for L51, L23, M269 and M343 (and the phyloequivalents of each)? If you think that's possible then I don't know what to say. All the Y-calls are being reassessed and we'll know the final answer when they finally publish the results. Besides, your Darra-i Kur sample is flagged as damaged in the excel file from the paper.
What I or other forum members think? Try what everyone is thinking. Everyone is asking the same questions about this study because the SNP calls are inconsistent. Manually checking it is the best bet right now, and people have manually checked it (see my quote of Angela quoting Megalophias). We'll know more when this is past the pre-print stage and finalized.
https://mobile.twitter.com/RohlfsenChad/status/980200022284357632
I got curious and here is some data of Darra-i Kur and I8998 (aka S8998) from the Ycall file that one of the authors uploaded to Dropbox:
Ancestral means - (negative),
derived means +(positive)
Sample ID Haplogroup Ancestral Derived
Darra-i Kur is some kind of BT, the sample is damaged (as per another supplementary excel file) so who knows what the real Y-hg is. Hell, it's not even clear what haplogroup it is, we just know that he is probably something under BT, which is no surprise, since that's a lot of people. With the way phylogeny works these SNP calls contradict each other at a fundamental level, how can you be positive for a mutation but have multiple ancestral calls for upstream markers? It's not possible, for example you can't be I2a but negative for I2, I, AND IJ but positive for haplogroup R mutations. Like getting 100% on a test but getting half of the questions wrong, it's a mistake. In this case it is a mistake made by a new software they used.
Darra-I Kur is a damaged sample and that is even stated in the paper’s supplementary table also the amount of SNP coverage (36,481 << that’s pretty low coverage) is poor and reflects the damage.
Numbers on left are ancestral (-) and numbers on far right column are (+)
Darra.I.Kur_d A00 1 0
Darra.I.Kur_d A0-T 0 2
Darra.I.Kur_d BT 0 1
Darra.I.Kur_d A1b1a1 1 0
Darra.I.Kur_d DE 1 0
Darra.I.Kur_d A1b1b2b 1 0
Darra.I.Kur_d E 1 0
Darra.I.Kur_d D 1 0
Darra.I.Kur_d C 1 0
Darra.I.Kur_d E2 1 0
Darra.I.Kur_d C2 1 0
Darra.I.Kur_d E1a 1 0
Darra.I.Kur_d E2b 2 0
Darra.I.Kur_d C2b 2 0
Darra.I.Kur_d G 1 0
Darra.I.Kur_d E1a2 1 0
Darra.I.Kur_d E2b1 1 0
Darra.I.Kur_d C2e1 1 0
Darra.I.Kur_d E1a2b 1 0
Darra.I.Kur_d C2e1a 1 0
Darra.I.Kur_d G1b 1 0
Darra.I.Kur_d G2a 2 0
Darra.I.Kur_d H3 5 0
Darra.I.Kur_d E1a2a2 2 0
Darra.I.Kur_d E2b1a1 1 0
Darra.I.Kur_d C1a1a2 1 0
Darra.I.Kur_d H3a 1 0
Darra.I.Kur_d H3b 1 0
Darra.I.Kur_d I 1 0
Darra.I.Kur_d LT/K1 1 0
Darra.I.Kur_d C2b1a2a 1 0
Darra.I.Kur_d C2e1a1a 1 0
Darra.I.Kur_d H1b2 1 0
Darra.I.Kur_d H3b1 3 0
Darra.I.Kur_d E1b1b1b1 1 0
Darra.I.Kur_d C1b1a1a1 1 0
Darra.I.Kur_d H1b2a 1 0
Darra.I.Kur_d H3a2a 1 0
Darra.I.Kur_d H3a2b 2 0
Darra.I.Kur_d E1b1a1a1a 1 0
Darra.I.Kur_d E1b1a1a1c 1 0
Darra.I.Kur_d I1a3 1 0
Darra.I.Kur_d I2a2 2 0
Darra.I.Kur_d E1a2b1a1a1 1 0
Darra.I.Kur_d E1b1b1a1b1 1 0
Darra.I.Kur_d C1b1a1a1a1a 1 0
Darra.I.Kur_d H1a1d2c2 1 0
Darra.I.Kur_d O1a 1 0
Darra.I.Kur_d O2a 1 0
Darra.I.Kur_d E1b1a1a1c2b2 1 0
Darra.I.Kur_d G2a2a1a2a1 1 0
Darra.I.Kur_d G2a2b2b1a1 1 0
Darra.I.Kur_d H1a1d2b2a 1 0
Darra.I.Kur_d I1a2a1a 1 0
Darra.I.Kur_d O2a1 1 0
Darra.I.Kur_d Q1b 2 0
Darra.I.Kur_d E1b1a1a1c2c1a 2 0
Darra.I.Kur_d H1a1d2c1a1 1 0
Darra.I.Kur_d O2a2b 1 0
Darra.I.Kur_d Q1b2 1 0
Darra.I.Kur_d G2a2b2a1a1c2 1 0
Darra.I.Kur_d T1a3b2a1 1 0
Darra.I.Kur_d O1b1a1 1 0
Darra.I.Kur_d Q1a2c 1 0
Darra.I.Kur_d Q1b2a 1 0
Darra.I.Kur_d G2a2b2a1a1a2a 1 0
Darra.I.Kur_d O1a1a1a 1 0
Darra.I.Kur_d O2a2b2a 1 0
Darra.I.Kur_d G2a2b2a1a1c1a1a1 1 0
Darra.I.Kur_d O1b1a1a1a2 1 0
Darra.I.Kur_d O1b1a1a1b1a 1 0
Darra.I.Kur_d O2a2b1a1a6b 1 0
Darra.I.Kur_d R1a1a1b2a2a 1 0
Darra.I.Kur_d R1b1a1a2a1a 0 1 (1 derived, no corroborating calls for upstream)
Darra.I.Kur_d R1b1a1a2a1a1c3 1 0
Darra.I.Kur_d R1b1a1a2a1a2a3 1 0
Darra.I.Kur_d R1b1a2a1a1b1a1a 1 0
Darra.I.Kur_d R1b1a1a2a1a1c2b3a 1 0
Darra.I.Kur_d R1b1a1a2a1a2c1g1b1 1 0
Darra.I.Kur_d R1b1a1a2a1a1c2b2a1b1a 1 0
This guy (is I8998 Swat IA) he is most definitely not U106. He has 25 derived calls for R, one derived call for R2. He's probably some kind of R, maybe R2. He is definitely not R-U106, and again we'll know the full story when they finalize the paper and share their haplogroup corrections, which has actually happened. We already know their Y-calls were an issue.
ID HG ANCESTRAL DERIVED
S8998.E1.L1 A00 36 0
S8998.E1.L1 A0-T 0 12
S8998.E1.L1 A0 16 0
S8998.E1.L1 A1 0 6
S8998.E1.L1 A0a 2 0
S8998.E1.L1 A0b 12 0
S8998.E1.L1 A1a 8 0
S8998.E1.L1 A1b 0 4
S8998.E1.L1 A0a1 9 1
S8998.E1.L1 A0a2 4 0
S8998.E1.L1 A1b1 1 0
S8998.E1.L1 BT 1 194
S8998.E1.L1 CT 1 104
S8998.E1.L1 DE 15 0
S8998.E1.L1 CF 0 2
S8998.E1.L1 E 74 0
S8998.E1.L1 D 25 0
S8998.E1.L1 C 79 1
S8998.E1.L1 F 0 13
S8998.E1.L1 D1 1 0
S8998.E1.L1 D2 3 0
S8998.E1.L1 GHIJK 0 1
S8998.E1.L1 D1b 3 0
S8998.E1.L1 G 93 0
S8998.E1.L1 HIJK 0 1
S8998.E1.L1 D1b2 19 0
S8998.E1.L1 H 17 0
S8998.E1.L1 IJK 0 1
S8998.E1.L1 D1a1a 1 0
S8998.E1.L1 D1a2a 1 0
S8998.E1.L1 D1b1a 1 0
S8998.E1.L1 D1b2a 19 0
S8998.E1.L1 H1 1 0
S8998.E1.L1 H2 4 0
S8998.E1.L1 H3 88 0
S8998.E1.L1 IJ 6 0
S8998.E1.L1 K 0 1
S8998.E1.L1 D1a2a1 1 0
S8998.E1.L1 D1b1a2 1 0
S8998.E1.L1 D1b1d1 20 0
S8998.E1.L1 D1b2a1 1 0
S8998.E1.L1 D1b2a2 4 0
S8998.E1.L1 I 57 0
S8998.E1.L1 J 63 1
S8998.E1.L1 LT/K1 10 0
S8998.E1.L1 K2 0 1
S8998.E1.L1 D1b1a2a 1 0
S8998.E1.L1 D1b1d1a 46 0
S8998.E1.L1 D1b1d1b 5 0
S8998.E1.L1 H1a1 3 0
S8998.E1.L1 H1a2 1 0
S8998.E1.L1 H1b1 103 0
S8998.E1.L1 H1b2 20 0
S8998.E1.L1 L/K1a 36 1
S8998.E1.L1 T/K1b 30 0
S8998.E1.L1 NO/K2a 4 0
S8998.E1.L1 K2c 2 0
S8998.E1.L1 K2d 1 0
S8998.E1.L1 D1b1a2b1 4 0
S8998.E1.L1 H1a1d 1 0
S8998.E1.L1 H1a2a 7 1
S8998.E1.L1 H1b2a 43 0
S8998.E1.L1 NO1 2 0
S8998.E1.L1 K2b1 1 0
S8998.E1.L1 P/K2b2 0 1
S8998.E1.L1 D1b1a2b1a 4 0
S8998.E1.L1 H1a1d1 43 1
S8998.E1.L1 H1a1d2 43 0
S8998.E1.L1 H1a2a1 42 0
S8998.E1.L1 T1a 2 0
S8998.E1.L1 N/K2a1 1 0
S8998.E1.L1 O/K2a2 28 0
S8998.E1.L1 K2b1b 1 0
S8998.E1.L1 M/K2b1d 1 0
S8998.E1.L1 P1/K2b2a 0 9
S8998.E1.L1 T1a1 7 0
S8998.E1.L1 T1a2 1 0
S8998.E1.L1 N1 1 0
S8998.E1.L1 O1 2 0
S8998.E1.L1 O2 5 0
S8998.E1.L1 K2b1a1 3 0
S8998.E1.L1 K2b1a2 1 0
S8998.E1.L1 K2b1a3 1 0
S8998.E1.L1 S/K2b1a4 2 0
S8998.E1.L1 M1 2 0
S8998.E1.L1 M2 1 0
S8998.E1.L1 Q/K2b2a1 10 0
S8998.E1.L1 R/K2b2a2 0 25 (some kind of R)
S8998.E1.L1 R1 12 0
S8998.E1.L1 R2 0 1<<< this makes more sense
S8998.E1.L1 R1a 3 0
S8998.E1.L1 R1b 1 0(1 ancestral, 0 derived)
S8998.E1.L1 R2a 3 0
S8998.E1.L1 R1a1 1 0
S8998.E1.L1 R1b1 6 1(6 ancestral, 1 derived)
S8998.E1.L1 R2a3 1 0
S8998.E1.L1 R1a1a 1 0
S8998.E1.L1 R1b1a 10 0(10 ancestral, 0 derived)
S8998.E1.L1 R1b1c 3 0
S8998.E1.L1 R2a3a 3 0
S8998.E1.L1 R1a1a1 2 0
S8998.E1.L1 R1b1a2 2 0
S8998.E1.L1 R2a3a1 1 0
S8998.E1.L1 R1a1a1b 2 0
S8998.E1.L1 R1b1a1a 11 0(11 ancestral calls)
S8998.E1.L1 R1b1a2b 1 0
S8998.E1.L1 R1b1a1a1 1 0(ancestral)
S8998.E1.L1 R1b1a1a2 17 1(17 ancestral, 1 derived)
S8998.E1.L1 R2a3a2b2 1 0
S8998.E1.L1 R2a3a2b2b 1 0
S8998.E1.L1 R2a3a2b2c 1 0
S8998.E1.L1 R1a1a1b1a3 1 0
S8998.E1.L1 R1b1a1a2a1 1 0
S8998.E1.L1 R1b1a1a2a2 2 0
S8998.E1.L1 R2a3a2b2b1 1 0
S8998.E1.L1 R1a1a1b1a1b 1 0
S8998.E1.L1 R1a1a1b1a2b 1 0
S8998.E1.L1 R1a1a1b1a3a 1 0
S8998.E1.L1 R1a1a1b1a3b 1 0
S8998.E1.L1 R1a1a1b2a1b 1 0
S8998.E1.L1 R1a1a1b2a2a 1 0
S8998.E1.L1 R1a1a1b2a2b 1 0
S8998.E1.L1 R1b1a1a2a1a 3 0
S8998.E1.L1 R1b1a1a2a2a 1 0
S8998.E1.L1 R2a3a2b2a1a 1 0
S8998.E1.L1 R1a1a1b1a1b1 1 0
As you can see there are a lot of contradictions in these SNP calls (false positives, etc). How can a haplogroup such as R-P311 be assigned when there are no supporting derived calls upstream of P311? How can one be assigned R-U106 when they are clearly negative for upstream SNPs? It really looks like Megalophias and several others who've looked at this already are correct in their assessment of the Ycall excel sheet from the paper provided via Dropbox. The Y-calls contradict the haplogroup assignments here, no question. Hell, even MA1 (Ma'alta boy) was given a weird call, when we already know he is just R*. Anyway, this is all we can say now until the raw data is released and then people can really dig into the SNP calls.
Hell, I have even looked at their calls for the I2a2a guys and they were negative for upstreams like IJ, I-M170, I2a2, etc. The Y-calls really need to be reassessed for this study, and they are working on that.
You cannot say R1a and R1b then R1a-Z95. The R1a from all Iran and Middle-East and Western Iran will mostly be R1a-Z93 wich Z95 is a downstream. As for R1b, its mostly Z2105 + minor lineages as V1636, for wich we all have almost a story. The R1a-R1b hybrid branches in Germanic countries are not Z93 nor Z2103.
Indeed, but here we go....
The South and Central Asian paper is a complete shit show with a lot of weird calls, dating and assumptions. Its been postponed for a lot of times now and is probably not a good argument until the complete paper is out and that people have put their hands on the samples. As exemple, the preprint of the paper came 2 months before the preprint for the Caucasus paper, wich was out a few months earlier on.
Agreed, I hope when they finally publish it with everything revised and finalized we'll finally be totally confident in the results.
The R1a-R1b hybrid branches in Iran is not Z93 nor Z2103 too, both of them and their subclades don't exist in Iran, even 0.001℅.
You sure about that?
Underhill et. al. (2015)
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4266736/#__sec7title
In the complementary R1a-Z93 haplogroup, the paragroup R1a-Z93* (Figure 3b) is most common (>30%) in the South Siberian Altai region of Russia, but it also occurs in Kyrgyzstan (6%) and in all Iranian populations (1–8%). R1a-Z2125 (Figure 3c) occurs at highest frequencies in Kyrgyzstan and in Afghan Pashtuns (>40%). We also observed it at greater than 10% frequency in other Afghan ethnic groups and in some populations in the Caucasus and Iran. Notably, R1a-M780 (Figure 3d) occurs at high frequency in South Asia: India, Pakistan, Afghanistan, and the Himalayas. The group also occurs at >3% in some Iranian populations and is present at >30% in Roma from Croatia and Hungary, consistent with previous studies reporting the presence of R1a-Z93 in Roma.31, 51Finally, the rare R1a-M560 was only observed in four samples: two Burushaski speakers from north Pakistan, one Hazara from Afghanistan, and one Iranian Azeri. - Underhill et al (2015)
R-Z93 subclades are found amongst Iranian-speaking populations. Even if the rate is low in Iran, it is still there. Again though, there was a population already present in Iran prior to the arrival of Iranian speakers. We already know J2 is the most common haplogroup in Iran, it seems more likely to me at least that we have an Iranian-speaking elite with a native substrate, we see this in a lot of places in history, not a new phenomenon. The elite-dominance mechanism is seen in a lot of places.
In response to you saying that there is no Z2103…
to quote Grugni et. al. (2012):
Eighty-eight Y-chromosome binary genetic markers were hierarchically genotyped as AFLP (YAP, [30]), RFLP (M2 [31], SRY10831.2 [32], M12 [33]; P15 [34]; M74 [35]; M34, M60, M61, M67, M70, M76, M78, M81, M175, M198, M207, M213 [36]; LLY22g, P36.2, P43 [37]; M123, M172 [38]; M242, M253, M285 [23]; V12, V13, V22 [39]; M377 [24]; P128, P287 [40]; M406 [41]; M269 [42]; Page08 [43]; V88 [44]; M458 [45]; PAGE55 [46]; L23, M412 [47]; L91 [48]; M527, M547, Page19, P303, U1 [49]), by DHPLC (M217 [50]; M25, M35, M47, M68, M69, M82, M92, M124, M170, M173, M174, M201, M205, M214, M216 [36]; M429 [51]; P209 [40]; M241, M267, M343 [23]; M357, M378, M410 [24]; M346 [40]; M434, M458 [45]; M530 [46]; L497, P16 [49]), and direct sequencing (M18 [33]; M42, M73, M75, M96 [52]; M33, PN2 [36]; MEH2 [53]; M317 [24]; M356 [54]; M438 [51]; P297 [40]). - Grugni et al (2012)
Do you see in the list of tested SNPs “Z2103”? I don’t. It means they didn’t test for it.
R-L23* shown as 8.5% in list you just linked, however again, this study which we already know didn't test more than the SNPs listed didn't test for Z2103. Do you see why this study doesn't really tell us much? We can't see the specific breakdown of the clades. So how much of that L23 (barring those which are indeed L23*) are Z2103? This is the same study that we already debated about R1a because of basal offshoots, this could be remedied if samples could be retested for downstream deeper SNPs. I wish they would do that for the sake of refinement at least. Let's wait for the final publication of the Central and South Asian study before we using it as a reference in discussions just yet.
But go ahead, deny this.