PDA

View Full Version : What Y-DNA haplogroup will Oase 1 have?



Passa
06-16-2015, 07:29 PM
Meanwhile the study is about to be published, I thought it would have been nice to make some guesses. What's your choice?

Anabasis
06-16-2015, 07:46 PM
Any Abstract? And where is Oase and when it is dated?

Passa
06-16-2015, 07:49 PM
Any Abstract? And where is Oase and when it is dated?

Ann Gibbons, Science, Ancient DNA pinpoints Paleolithic liaison in Europe (http://www.sciencemag.org/content/348/6237/847.summary)

To Erik Trinkaus, the jaw of the oldest modern human found in Europe has always looked strange. Its huge wisdom teeth and hefty, buttressed lower jaw reminded him of Neandertals, and he argued that this fossil, 37,000 to 42,000 years old, was the product of generations of mixing between modern humans and our extinct cousins. “It wasn't a popular idea,” admits Trinkaus, a paleoanthropologist at Washington University in St. Louis. Other paleoanthropologists insisted that the young man whose remains were found in 2002 in Peştera cu Oase cave in Romania was just a chunky example of our own species.

Now, 15 years later, Trinkaus has been vindicated by ancient DNA. The young Oase man inherited as much as one-tenth of his DNA from a Neandertal ancestor, and that ancestor lived only 200 years or so previously, according to a talk this month at Cold Spring Harbor Laboratory in New York. “One of Oase's ancestors—its great-great-great-grandparent—is Neandertal,” reported Qiaomei Fu, a geneticist at the Chinese Academy of Sciences–Max Planck Society Joint Laboratory for Human Evolution in Beijing and a postdoc in the lab of population geneticist David Reich at Harvard Medical School. The finding is “important as the first direct evidence of a very recent admixture event in Europe,” says population geneticist Laurent Excoffier of the University of Bern.

Europe just after the arrival of modern humans has long seemed a likely setting for such close encounters, given that Neandertals and modern humans overlapped there about 45,000 to 39,000 years ago. But until now, ancient DNA pointed to a different time and place for such a liaison. By sequencing the genomes of fossil Neandertals and comparing them with today's human genomes, paleogeneticists had found that living Europeans and Asians—but not Africans—have inherited just 1% to 4% of their DNA from Neandertals. DNA from fossils of two modern humans from what is now Russia also suggested that their Neandertal heritage was faint (see http://scim.ag/RussDNA). So researchers proposed that modern humans and Neandertals had rare and relatively early encounters, perhaps in the Middle East, when moderns swept out of Africa 60,000 to 50,000 years ago.

The DNA from Oase 1, a lower jaw without a skull, complicates that picture, Fu reported at the Biology of Genomes meeting. Working in a team led by paleogeneticist Svante Pbo of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, she and her colleagues captured 2.2 million base pairs of the fossil's DNA. Then, they sequenced 78,055 locations where the genomes of Neandertals and modern humans are known to differ. They found that the Oase man had far more Neandertal DNA—composing 4.8% to 11.3% of his genome—than either the ancient modern humans from Russia or living Europeans and Asians, Fu said.

What's more, the young man had inherited the Neandertal DNA in “large chunks,” including several segments more than 50 million base pairs long; one chunk spanned half the length of chromosome 12. Those unbroken stretches of Neandertal DNA suggest that the interbreeding must have been just four to six generations back. If the mixing had been more ancient, the long DNA segments would have been broken up by the reshuffling of chromosomes that takes place every generation. “This is quite amazing,” Fu said in her talk. “We're quite excited about that.”

If modern humans and Neandertals had several successful matings, why do living humans' genomes record only the earlier event? An answer emerged when Fu traced how this Oase man and other early modern fossils link to later peoples. One of the early modern fossils from Russia, a 36,000- to 39,000-year-old arm bone known as Kostenki 14, is genetically similar to present-day Europeans. In contrast, the DNA of the Oase fossil, although it is from Europe, more closely resembles ancient Asians than Kostenki or living Europeans, Fu reported. The team concluded that the Oase man himself was an evolutionary dead end, who did not pass his DNA along to living Europeans. Members of the team have declined to comment further, because their report is in press.

Fu's was “an impressive talk,” says population geneticist Andrew Clark of Cornell University, and suggests that “there were many interbreeding events.” But the recent mixing surprises some. “I thought that interbreeding would be a lot less likely at 40,000 years when there were so few Neandertals left,” says Chris Stringer of the Natural History Museum in London. “I have to admit Erik has been proved right.”

Trinkaus wasn't surprised when he saw a copy of the manuscript. “It confirms things a bunch of us have been saying for a long time.”

J Man
06-16-2015, 08:35 PM
I am guessing C.

Passa
06-16-2015, 08:41 PM
I am guessing C.

Kostenki 14 (a C1a carrier) had European genetic material, so I don't think Oase 1 has this lineage. He is described as being similar to "ancient Asians" (Ust'-Ishim? Jomons?) and he did not leave any sign on modern European populations. I'm going with DE/D/E.

J Man
06-16-2015, 10:58 PM
Kostenki 14 (a C1a carrier) had European genetic material, so I don't think Oase 1 has this lineage. He is described as being similar to "ancient Asians" (Ust'-Ishim? Jomons?) and he did not leave any sign on modern European populations. I'm going with DE/D/E.

I highly doubt that he will be DE, D or E.

Passa
06-16-2015, 11:02 PM
I highly doubt that he will be DE, D or E.

Why so?

Megalophias
06-17-2015, 12:37 AM
I put C, but really he could be almost anything, even A or B. I suppose they would have let something slip by now if he had a Neanderthal lineage.

I doubt he will be D, but some kind of DE* or E* is certainly possible.

J Man
06-17-2015, 01:23 AM
Why so?

Because I do not believe any of those haplogroups ( I am thinking E here mainly) were in Europe until the Neolithic or at the very earliest Mesolithic.

Anabasis
06-17-2015, 07:57 AM
I voted for C but IJ is also possible.

Megalophias
06-17-2015, 08:20 AM
Because I do not believe any of those haplogroups ( I am thinking E here mainly) were in Europe until the Neolithic or at the very earliest Mesolithic.

The E haplogroups that exist in Europe *now* weren't there until the Mesolithic or later, most likely. But here we are surely talking about an extinct lineage.

Passa
06-17-2015, 09:39 AM
Because I do not believe any of those haplogroups ( I am thinking E here mainly) were in Europe until the Neolithic or at the very earliest Mesolithic.

It is absurd to think that a 66-kyo haplogroup was never able to reach Europe before 9 kya.

Passa
06-17-2015, 09:40 AM
The E haplogroups that exist in Europe *now* weren't there until the Mesolithic or later, most likely. But here we are surely talking about an extinct lineage.

That's what I mean too.

J Man
06-17-2015, 10:17 AM
It is absurd to think that a 66-kyo haplogroup was never able to reach Europe before 9 kya.

Nothing absurd about it. I just do not think E was in Europe until the Neolithic basically. Each to his own.

Coldmountains
06-17-2015, 10:19 AM
Why not D? I also think his haplogrouo was most likely C but there are some few Europeans carrying haplogroup D so i wonder where it is from.

Coldmountains
06-17-2015, 10:21 AM
Bump.

Passa
06-17-2015, 10:37 AM
If the guy really didn't contribute to the modern European gene pool, he won't have the following Eurogenes K15 components: North_Sea, Atlantic, Baltic, Eastern_Euro, West_Med, West_Asian, East_Med. Excluding East_Asian et al. components and also African components, we are left with the following possible components: Red_Sea, South_Asian, Oceanian.

T101
06-17-2015, 10:50 AM
haha all in on C!

DMXX
06-17-2015, 11:14 AM
Going for Y-DNA F, some sort of extinct line (given Passa's comment about a lack of contribution to modern Europeans). No empirical justifications for it, more a hunch than anything else.

parasar
06-17-2015, 12:35 PM
Other.
Neandertal Y line.

DMXX
06-17-2015, 12:39 PM
Other.
Neandertal Y line.

If heads turned super-swiftly at the Euro HG Y-DNA C's, this will cause rotational force self-decapitations among many DNA enthusiasts. :D

Kale
06-17-2015, 01:02 PM
^ Ha. I'd like to see that.

Anyways...my vote is on F*. Kind of dull, but that's where the hunch is. I think pretty much anything else would be more interesting.
A - There's some random A1a* in Finland, or so I hear, would spark a lot more investigation of A in Europe + elsewhere, instead of just dismissal as African.
B - how would a pygmy haplogroup end up there?...how funny it would be to see the 'modern european = pygmy + neanderthal' jokes.
C - Oase is more directly from the Middle East, whereas Kostenki and others could be some sort of Central Asian offshoot...a C here would actually be interesting, although likely, and perhaps expected.
D - I don't even know how you'd explain that...
E - Would perhaps further promote a Eurasian origin for it
F -
G - It formed 50,000 years ago but doesn't coalesce until 30,000 years ago...where was it during that time?
H,K1 - That would be weird, although not impossible, we do have that one odd H in the neolithic
I or J - There goes your gravettian hypothesis?
K*/K2 - I suppose that's also reasonable...would definitely clash with the Southeast Asian origin that's been so prominent as of late.

Passa
06-17-2015, 01:13 PM
D - I don't even know how you'd explain that...
E - Would perhaps further promote a Eurasian origin for it


I explain D in this way. D is a very old haplogroup with a very wide distribution (I saw an Irish with this haplogroup and also a Basque), so an extinct sub-clade could have made it to Europe. Plus, the article says that Oase 1 resembled "ancient Asians". Since the author of the article mentioned Kostenki 14 but not Ust'-Ishim, I imagine that these "ancient Asians" might be the Jomon genomes analysed by Kanzawa-Kiriyama. These ancient Jomons in fact are the most South-Asian-shifted (and therefore West-Eurasian-shifted) compared with modern East and Southeast Asians. This might mean that the original, unmixed ancestors of the Jomons may have very well been South-Asian-like (see Onge and Jarawa, 100% Y-DNA D, and 100% South Asian autosomally). I apply the same to Y-DNA DE, which was originally Eurasian (and originated in between Syria and Pakistan IMO).

Regarding E, it is 66 kyo and originated most likely around the Levant/SW Asia (Saudi Arabia has the highest concentration of E* in the world). So it is a likely candidate too.

Passa
06-17-2015, 01:32 PM
This is what I was referring to in the previous post regarding D.

4919

Kale
06-17-2015, 02:56 PM
This is what I was referring to in the previous post regarding D.

4919

Papuans often occupy a similar position on PCAs, but with a slight African shift as well (perhaps reflecting their Denisovan ancestry?)...I imagine Onge would plot where Jomon do as well.

Passa
06-17-2015, 03:02 PM
Papuans often occupy a similar position on PCAs, but with a slight African shift as well (perhaps reflecting their Denisovan ancestry?)...I imagine Onge would plot where Jomon do as well.

Onge are Australoid-shifted South Asians in the plot at page 161 here: http://www.sciencemag.org/content/suppl/2011/09/21/science.1211177.DC1/Rasmussen.SOM.pdf

Kale
06-17-2015, 04:02 PM
Or South Asians are West Eurasian shifted Australoids? :P

parasar
06-22-2015, 03:22 PM
"This allowed us to define Oase 1 as belonging to macrohaplogroup F
(positions given in hg19 coordinates)."
http://www.nature.com/nature/journal/vaop/ncurrent/extref/nature14558-s1.pdf

RCO
06-22-2015, 05:36 PM
I voted F. I thought at that age haplogroup C would still be expanding far away to the East-Orient, only some C branches moved later to Europe via Siberia-Central Asia and F branches would be around the Western Asian, South-Eastern European spaces of GHIJK, specially the European spaces later occupied by F branches like I because P-Q-R would had the same long Eastern trail of C returning later to Eastern Europe via Siberia-Central Asia.

Kale
06-22-2015, 05:43 PM
^ Exactly my reasoning as well. I'm more surprised about the basal mtdna N than y-hg F.

J Man
06-22-2015, 06:11 PM
I voted C but F makes a lot of sense as well.

parasar
06-22-2015, 06:19 PM
I voted F. I thought at that age haplogroup C would still be expanding far away to the East-Orient, only some C branches moved later to Europe via Siberia-Central Asia and F branches would be around the Western Asian, South-Eastern European spaces of GHIJK, specially the European spaces later occupied by F branches like I because P-Q-R would had the same long Eastern trail of C returning later to Eastern Europe via Siberia-Central Asia.

But we have to keep in mind that while Oase 1 (Y-F, ~40000ybp) and Kostenki 14 (Y-C, ~39000ybp) are both from about the same time-frame, the former has lower affinity to modern Europeans (and likely to west Eurasians) when compared to east Eurasians, while the latter is closer to Europeans.

R.Rocca
06-22-2015, 06:33 PM
I voted C but F makes a lot of sense as well.

^^^ me too.

genetiker
06-25-2015, 12:54 AM
He belonged to K:

https://genetiker.wordpress.com/2015/06/24/analyses-of-the-oase-1-genome/

parasar
06-25-2015, 01:28 AM
He belonged to K:

https://genetiker.wordpress.com/2015/06/24/analyses-of-the-oase-1-genome/

Thanks.
This makes more sense as I was having a tough time reconciling his autosomal data with an F designation (post #32 above).


The Y-SNP calls for Oase 1 show that he belonged to Y haplogroup K*. He had a positive call for the SNP PF3500 at position 21571895, which is at the level of IJK in the Y chromosome tree. He also had a positive call for the SNP PF5495 at position 15842844, which is at the level of K. He had negative calls for SNPs defining haplogroups LT, P, N, and O.

Like the 45,000-year-old Ust-Ishim sample from Siberia, and the 40,000-year-old Tianyuan sample from China, the largest autosomal components for Oase 1 were the Veddoid South Asian or Indian components. This once again confirms the account of extra-African human evolution that I introduced two and a half years ago, in which Negritoids in India evolved into Veddoids, and Veddoids then evolved into Caucasoids and Mongoloids.

parasar
06-25-2015, 01:33 AM
http://www.yfull.com/share/yreport/4c2c80e4062c5a7a286d5539400de0ac/

IJK L16 • M522 • S138 • PF3493 V1438 • YSC0001319 • PF3497 • M2684 L15 • M523 • S137 • PF3492 • Z4413 PF3500 • M2696 V1295 • PF3495 • F3689 • M2682 • Y2571

K P132 • PF5480 PF5500 • F2548 • M2692 L819 • CTS4265 • PF5494 • M2686 V3169 • PF5495 • F2006 • M2688 PF5470 A5331 • YSC0000055 • PF5459 • M2348 M2352 • CTS2071 • PF5489 P131 • PF5493 L469 • PF5499 • M2689 PF5469 • V104 M9 • PF5506 PF5488 • M2351 V4038 • PF5503 • F3026 • M2694 P128 • PF5504 YSC0000222 • PF5505 • L1346 • M2695 CTS9278 • PF5501 • M2693 CTS10976 • PF5509 • M2698FF

vettor
06-25-2015, 01:40 AM
LOL
Champagne for Sweuro and myself:beerchug:

cheers Sweuro:)

lgmayka
06-25-2015, 01:43 AM
He belonged to K:

https://genetiker.wordpress.com/2015/06/24/analyses-of-the-oase-1-genome/
According to that list of positive SNP results, he belongs more specifically to NO. According to YFull, NO is about 41,000 years old (http://yfull.com/tree/NO/), but some people claim that YFull underestimates by 10-20%.

However, the location that Genetiker uses for NO, 6753303, is suspect. Known as L76 or M8964, YFull says that it is found in dozens of places all over the haplotree. Genetiker needs to check this sample for the 48 reliable SNPs associated with NO:

Z4901 • M2318 • S16324 CTS8891 F415 • M2327 • S19886 Z4843 • M2309 CTS9599 • M2333 Z4896 F176 • M2314 Y2158 S8850 P192 Z4817 Z4942 • M2337 • K482 Z4965 • M2341 CTS11623 CTS3544 • M2319 CTS5127 • M2323 CTS7029 • M2326 F7 • M2304 • S8255 Z4748 Z4866 • M2315 CTS8274 • M2331 CTS3873 Z4953 • M2340 CTS5858 • M2325 CTS7747 • M2328 M2311 Z4944 • M2338 P195 CTS11012 • M2342 CTS2575 • M2317 CTS9775 • M2334 CTS7905 • M2330 Z4775 F306 • M2322 • S17726 Z4940 • M2336 P193 Z5020 PAGES00039 • M214 • PAGE39 P194 F107 • M2310 • S10947 F287 • M2320 F468 • M2332 • S20648 CTS11572 CTS4135 • M2321 F53 • M2305 • S9681 CTS7748 • M2329 L838 • CTS11536 • M2344 F658 • M2347 • S25824

malleus
06-25-2015, 02:49 AM
I doubt genetiker's result because he is negative at 6994764. The list suggests that the frequency of false positives is low but much much higher than for modern samples while false negatives are extremely rare. This suggests that it is more likely that at 15842844 it has a false positive rather than 6994764 is a false negative. The last reliabe branch is FK(whatever that means) which has 11 consecutive positives without a negative. The researchers had a good reason to be cautious when they announced that Oase belonged to F*. A third possibility also exists; both 15842844 and 6994764 are good calls and he belongs to a K lineage that diverged before LMNOPT.

Motzart
06-25-2015, 05:34 AM
According to that list of positive SNP results, he belongs more specifically to NO. According to YFull, NO is about 41,000 years old (http://yfull.com/tree/NO/), but some people claim that YFull underestimates by 10-20%.

However, the location that Genetiker uses for NO, 6753303, is suspect. Known as L76 or M8964, YFull says that it is found in dozens of places all over the haplotree. Genetiker needs to check this sample for the 48 reliable SNPs associated with NO:

Z4901 • M2318 • S16324 CTS8891 F415 • M2327 • S19886 Z4843 • M2309 CTS9599 • M2333 Z4896 F176 • M2314 Y2158 S8850 P192 Z4817 Z4942 • M2337 • K482 Z4965 • M2341 CTS11623 CTS3544 • M2319 CTS5127 • M2323 CTS7029 • M2326 F7 • M2304 • S8255 Z4748 Z4866 • M2315 CTS8274 • M2331 CTS3873 Z4953 • M2340 CTS5858 • M2325 CTS7747 • M2328 M2311 Z4944 • M2338 P195 CTS11012 • M2342 CTS2575 • M2317 CTS9775 • M2334 CTS7905 • M2330 Z4775 F306 • M2322 • S17726 Z4940 • M2336 P193 Z5020 PAGES00039 • M214 • PAGE39 P194 F107 • M2310 • S10947 F287 • M2320 F468 • M2332 • S20648 CTS11572 CTS4135 • M2321 F53 • M2305 • S9681 CTS7748 • M2329 L838 • CTS11536 • M2344 F658 • M2347 • S25824

Thats consistent with the high levels of east asian admixture in the autosomal analysis.

genetiker
06-25-2015, 05:54 AM
According to that list of positive SNP results, he belongs more specifically to NO. According to YFull, NO is about 41,000 years old (http://yfull.com/tree/NO/), but some people claim that YFull underestimates by 10-20%.

However, the location that Genetiker uses for NO, 6753303, is suspect. Known as L76 or M8964, YFull says that it is found in dozens of places all over the haplotree.

L76/M8964 is not actually at the level of NO. It's positive in R genomes and negative in I and J genomes. So it's also at the level of K.

Besides, the idea that a Paleolithic European was NO is nonsensical.

genetiker
06-25-2015, 05:56 AM
I doubt genetiker's result because he is negative at 6994764. The list suggests that the frequency of false positives is low but much much higher than for modern samples while false negatives are extremely rare. This suggests that it is more likely that at 15842844 it has a false positive rather than 6994764 is a false negative. The last reliabe branch is FK(whatever that means) which has 11 consecutive positives without a negative. The researchers had a good reason to be cautious when they announced that Oase belonged to F*. A third possibility also exists; both 15842844 and 6994764 are good calls and he belongs to a K lineage that diverged before LMNOPT.

The SNP at 6994764 is CTS753/PF5525, which is actually one of the defining SNPs for LT, not K.

malleus
06-25-2015, 06:37 AM
Besides, the idea that a Paleolithic European was NO is nonsensical.

Why? Ust Ishim was an early offshoot of K2a.
I noticed that some of "false positives" may actually be recurrent mutations.
In that case(if false results are rare) this guy has a very good chance of being K*.
But if he belongs to any of the known sub-branches of K, his line probably branched off quite early.
Currently he can be anything, NO*(K2a*) L* T* LT*(K1*) K2* K2b* P*(but not R* or Q*) etc. and of course K*.(TMRCA of NO is too close but N* or O* may be included)

Addendum

L76/M8964 is not actually at the level of NO. It's positive in R genomes and negative in I and J genomes. So it's also at the level of K.


Actually it is one of the defining SNPs for K2a(K-M2335 in YFull) and it is in the most upstream position among those consistent with its probably being within K.
The R genome that has it is in the downstream position. For instance Q does not have it, which means it cannot be a P* marker and this eliminates the posibility this guy is an early offshoot from P* or even R2b*.
So as much as this marker is unreliable, as it is, I would rate this guy's being K2a(NO*) a bit more likely than other posibilities.


The SNP at 6994764 is CTS753/PF5525, which is actually one of the defining SNPs for LT, not K.

Indeed. You made an error in your list and I just went by your list as I had no reason to check each one of them. Good work though otherwise. Where did you get the raw file?

Ebizur
06-25-2015, 08:56 AM
According to that list of positive SNP results, he belongs more specifically to NO. According to YFull, NO is about 41,000 years old (http://yfull.com/tree/NO/), but some people claim that YFull underestimates by 10-20%.Hallast et al. (2014) have given a point estimate of 30,000 years for the TMRCA of the NO-M214 clade and a point estimate of 32,600 years for the TMRCA of the K-M9 clade. Fu et al. (2014) have given a point estimate of 50,000 years for the TMRCA of the KxLT node (i.e. the TMRCA of all K2 Y-DNA) and a point estimate of 54,000 years for the TMRCA of the IJK node. Since the TMRCA estimates of Fu et al. are more consistent with the dating of ancient specimens, I think it is reasonable to adjust the TMRCA estimates of Hallast et al. on the basis of the TMRCA estimates of Fu et al.

The TMRCA of IJK sets an upper bound for the TMRCA of K-M9, and the TMRCA of K2 sets a lower bound for the TMRCA of K-M9. Taking the upper bound, the TMRCA estimate for K-M9 provided by Hallast et al. must be adjusted by a factor of 54,000/32,600 = 1.65644172. Taking the lower bound, the TMRCA estimate for K-M9 provided by Hallast et al. must be adjusted by a factor of 50,000/32,600 = 1.53374233.

Adjusting the estimate of the TMRCA of NO-M214 provided by Hallast et al. by the same correction factors results in an upper bound of 49,693 years (assuming the TMRCA of K-M9 is approximately equal to the TMRCA of IJK) and a lower bound of 46,012 years (assuming the TMRCA of K-M9 is approximately equal to the TMRCA of K2). My SNP-counting method using the data provided by Hallast et al. also has produced an estimate of the TMRCA of (N-M231 + O-P191) that falls between those two values. I would say it is quite likely that the MRCA of N-M231 and O-P191 has lived at some time between 45,000 and 50,000 years ago.

In conclusion, I think that those people who have claimed that YFull underestimates TMRCA by 10-20% are probably correct.

lgmayka
06-25-2015, 09:08 AM
L76/M8964 is not actually at the level of NO. It's positive in R genomes and negative in I and J genomes. So it's also at the level of K.
L76 is not nearly so simple as that--it appears many times (independently) all over the haplotree. Go to YFull's haplotree (http://yfull.com/tree/), and click on the Search button in the upper right corner. In the popup window, enter L76 and click the Search button.

lgmayka
06-25-2015, 09:11 AM
Besides, the idea that a Paleolithic European was NO is nonsensical.
Modern N is divided into two immediate subclades, N-Y6503 and N-Z4762 (http://yfull.com/tree/N/). N-Y6503 has never been found outside of Europe. In fact, it most commonly occurs in the Balkans.

Please check the sample's data for at least some of the 48 reliable SNPs defining NO.

Ebizur
06-25-2015, 09:22 AM
Has the Oase 1 specimen been found to be negative for any marker of haplogroup NO or any marker of haplogroup O? If not, I would not exclude that he might be downstream of the NO node (i.e. pre-N, pre-O, or some extinct sort of NO(xN, O)) or even downstream of the O node (i.e. pre-O1'2, pre-O3, or some extinct sort of O(xO1'2, O3)). The TMRCA of haplogroup O (O1'2 + O3) should be between roughly 39,000 and 42,500 years, which potentially would allow descendants of haplogroup O to already have diverged by the time of the Oase 1 specimen (dated to 37,000 ~ 42,000 YBP).

malleus
06-25-2015, 09:41 AM
While L76/M8964 seems recurrent in the entire Y tree, K2a(that includes NO) is the only possible place that is consistent with it being within K and without multiple recurrent mutations in the SAME lineage.
So as long as YFull tree was not grossly messed up and the L76/M8964 result is a valid call for this guy, the case is fairly strong.

And as I said he can even be N* or O* or even more downstream depending on the age estimates.

malleus
06-25-2015, 09:52 AM
Has the Oase 1 specimen been found to be negative for any marker of haplogroup NO or any marker of haplogroup O?

He is negative for 9 N* markers and positive for 1/12 O* markers. So if there is no false positive(or another recurrency) he appears to be a very early offshoot of O*.
Initially I did not even consider this possibility because of the apparent appalling rate of "false positives" but then realized false negatives seem very rare and theoretically they should occur at about the same rate as false positives.
Many of "false positives" may actually be recurrent mutations which won't be "false" in the literal sense.

He lived fairly close to the N-O split time so 1 out of 12 makes sense. On the other hand at that low frequency this may indeed be a false positive.

All this rests on the assumption that YFull tree made no error in placing SNPs. If you allow errors the only thing certain right now is that he is within F* as the original authors announced.

P.S. There is only 1 NO* marker that got a good call and it is, as mentioned before, positive. If one finds even a single NO* marker that is negative that means O-Y3287+ is recurrent or false. This is why I don't really feel confident of my claim.

Ebizur
06-25-2015, 10:19 AM
He is negative for 9 N* markers and positive for 1/12 O* markers. So if there is no false positive(or another recurrency) he appears to be a very early offshoot of O*.Do you mean that there are data available regarding the state of the Oase 1 specimen's Y-DNA at twelve loci relevant to the determination of haplogroup O, and that he is positive for one and negative for eleven of those twelve SNPs? In that case, he would be positioned very early on the pre-O branch (or on an extinct branch diverging from the pre-O branch immediately downstream of the NO node and upstream of the O node), and not anywhere downstream of the O node. In other words, he would not be a member of haplogroup O as it is commonly understood at present, though his phylogenetic position might be described as O2 if the present O is renamed to O1 (with the present O1'2 becoming O1a'b and the present O3 becoming O1c).

malleus
06-25-2015, 10:28 AM
Do you mean that there are data available regarding the state of the Oase 1 specimen's Y-DNA at twelve loci relevant to the determination of haplogroup O,...

It was from genetiker's site. Just click Y-SNP calls and you see the list. I asked him where he got the raw file but have not heard from him.
Yes he would be an O only in the extended sense(as I said "an early offshoot") just as Ust Ishim is a K2a in the extended sense.

I don't think you rename haplogroups to fit ancient samples. O1 and O2 will remain as they are and even if this result is genuine this guy will be just a side note just as Ust Ishim became.

P.S. I hate to speculate on things that are uncertain(I have low confidence that Oase is indeed NO* let alone O*), but if he were a modern sample the O clade would be renamed. He would be just O* under the new definition and the old O's would be renamed O-M**** or O-F**** depending on your favorite SNP system etc. Renaming along the line of O1 O2 is only done when each represents a significant population. Personally I hate it. I like the old O1 O2 O3 system intact.

malleus
06-25-2015, 03:00 PM
I tried to re-check Genetiker's list but it is too much work.
I have not found any additional error except the ones already mentioned.
1 K-6994764 should be LT*
2 KR-15842844 should be K* This position is equiv. to PF5495/F2006 etc. and YFull locates it in K*.

A There is only 1 valid IJK* call and it is positive.
B All IJ* markers are negative, also as mentioned in the actual paper.
C The lone valid K* marker is positive.

A+B+C makes a very strong case for Oase being within K.
There is only one "contiguous" marker under K with a positive call and it is NO.
Under NO there is also only 1 contiguous marker with a positive call and it is O.
So he has a reasonable chance of being O*.

By "contiguous", I meant there is no negative call between it and any upstream marker to it.
Using this criterion you can eliminate a lot of recurrency and false positives and as far as I checked IJK - > K - > NO - > O is the only line that satisfies this for Oase.

parasar
06-25-2015, 04:38 PM
I tried to re-check Genetiker's list but it is too much work.
I have not found any additional error except the ones already mentioned.
1 K-6994764 should be LT*
2 KR-15842844 should be K* This position is equiv. to PF5495/F2006 etc. and YFull locates it in K*.

A There is only 1 valid IJK* call and it is positive.
B All IJ* markers are negative, also as mentioned in the actual paper.
C The lone valid K* marker is positive.

A+B+C makes a very strong case for Oase being within K.
There is only one "contiguous" marker under K with a positive call and it is NO.
Under NO there is also only 1 contiguous marker with a positive call and it is O.
So he has a reasonable chance of being O*.

By "contiguous", I meant there is no negative call between it and any upstream marker to it.
Using this criterion you can eliminate a lot of recurrency and false positives and as far as I checked IJK - > K - > NO - > O is the only line that satisfies this for Oase.

If he is on the O line, one thing becomes clear - this line did not leave any descendants in that area.
"Haplogrupo O (ADN-Y) edit" by Maulucioni - Wikimedia Commons. Licensed under CC BY 3.0 via Wikimedia Commons"
https://upload.wikimedia.org/wikipedia/commons/2/23/Haplogrupo_O_%28ADN-Y%29_edit.png

Motzart
06-25-2015, 06:19 PM
East Asians do have the most Neanderthal Ancestry of any modern group. It is unfortunate that we have not yet seen an estimate of Neanderthal Ancestry in the Haak/Allentoft genomes.

vettor
06-25-2015, 06:26 PM
I tried to re-check Genetiker's list but it is too much work.
I have not found any additional error except the ones already mentioned.
1 K-6994764 should be LT*
2 KR-15842844 should be K* This position is equiv. to PF5495/F2006 etc. and YFull locates it in K*.

A There is only 1 valid IJK* call and it is positive.
B All IJ* markers are negative, also as mentioned in the actual paper.
C The lone valid K* marker is positive.

A+B+C makes a very strong case for Oase being within K.
There is only one "contiguous" marker under K with a positive call and it is NO.
Under NO there is also only 1 contiguous marker with a positive call and it is O.
So he has a reasonable chance of being O*.

By "contiguous", I meant there is no negative call between it and any upstream marker to it.
Using this criterion you can eliminate a lot of recurrency and false positives and as far as I checked IJK - > K - > NO - > O is the only line that satisfies this for Oase.

Any chance of he being the old X ydna ............IIRC , it sat between K and NO.

changed to:
Haplogroup K2e (K-M147) was previously known as ex-K2a and/or "Haplogroup X".


new tree split below
https://en.wikipedia.org/wiki/Paragroup

malleus
06-25-2015, 09:06 PM
Any chance of he being the old X ydna
new tree split below


That requires 2 recurrences or false positives. Not impossible but not likely either.
M147 is still unresolved as far as I know.

Ebizur
06-25-2015, 10:41 PM
If he is on the O line, one thing becomes clear - this line did not leave any descendants in that area.According to the data posted by malleus, Oase 1 would not belong to haplogroup O (i.e. he is not positive for all the mutations that all extant haplogroup O Y-chromosomes share). He would rather be a sort of pre-O (potentially the direct ancestor of extant O, but more likely on a side branch that is now extinct). His phylogenetic position would be analogous to that of Ust'-Ishim's Y-DNA vis--vis haplogroup X'NO or that of Mal'ta's Y-DNA vis--vis haplogroup R.

It is interesting that Oase 1 probably has lived roughly around the same time as the MRCA of all haplogroup O. What was the range of members of haplogroup NO at that time? Did they range all the way from the Balkans (Peștera cu Oase is located in southwestern Romania near the border with Serbia, and much closer to Belgrade than to Bucharest) to East Asia already, or did haplogroup O first appear in Europe before moving eastward and eventually all the way to southeastern Asia?

vettor
06-25-2015, 11:09 PM
According to the data posted by malleus, Oase 1 would not belong to haplogroup O (i.e. he is not positive for all the mutations that all extant haplogroup O Y-chromosomes share). He would rather be a sort of pre-O (potentially the direct ancestor of extant O, but more likely on a side branch that is now extinct). His phylogenetic position would be analogous to that of Ust'-Ishim's Y-DNA vis--vis haplogroup X'NO or that of Mal'ta's Y-DNA vis--vis haplogroup R.

It is interesting that Oase 1 probably has lived roughly around the same time as the MRCA of all haplogroup O. What was the range of members of haplogroup NO at that time? Did they range all the way from the Balkans (Peștera cu Oase is located in southwestern Romania near the border with Serbia, and much closer to Belgrade than to Bucharest) to East Asia already, or did haplogroup O first appear in Europe before moving eastward and eventually all the way to southeastern Asia?

pre O was called Xydna 6 months ago, now its k2e

basmaci
06-25-2015, 11:11 PM
N :party:

malleus
06-25-2015, 11:31 PM
pre O was called Xydna 6 months ago, now its k2e

X would be "pre-NO" if you have to use that terminology and I have yet to see an evidence that M147+ belongs to X.
Haplogroups are constantly redefined and expanded, otherwise you would even have to call Malta boy "pre-R".
I don't use "pre-" terminology since the definition of a haplogroup is quite arbitrary to begin with and is based on known populations.
As new results come in it is constantly revised and redefined. It has been that way for the last 15 years or so.

Kale
06-26-2015, 12:42 PM
I'm just going to say now...if Oase is haplogroup O!?!?!? Nobody can ever belittle the importance of lineage replacement/extinction again.

genetiker
07-04-2015, 04:02 AM
I've updated my list of Y-SNP calls for Oase 1. It should be clearer now.

From my revised post:

"The Y-SNP calls for Oase 1 show that he belonged to Y haplogroup K. He had positive calls for the IJK-defining SNPs PF3497 and PF3500, and for the K-defining SNP PF5495. Like Ust'-Ishim, Oase 1 had a positive call for the SNP Z4842 and a negative call for the SNP F650. These are two of the SNPs that define haplogroup X, which was created when the Y chromosome of an Indian Telugu was found to share seven mutations with haplogroup NO Y chromosomes. Like Ust'-Ishim, Oase 1 had only negative calls for NO-defining SNPs."

https://genetiker.wordpress.com/2015/06/24/analyses-of-the-oase-1-genome/

Krefter
07-04-2015, 05:53 AM
I've updated my list of Y-SNP calls for Oase 1. It should be clearer now.

From my revised post:

"The Y-SNP calls for Oase 1 show that he belonged to Y haplogroup K. He had positive calls for the IJK-defining SNPs PF3497 and PF3500, and for the K-defining SNP PF5495. Like Ust'-Ishim, Oase 1 had a positive call for the SNP Z4842 and a negative call for the SNP F650. These are two of the SNPs that define haplogroup X, which was created when the Y chromosome of an Indian Telugu was found to share seven mutations with haplogroup NO Y chromosomes. Like Ust'-Ishim, Oase 1 had only negative calls for NO-defining SNPs."

https://genetiker.wordpress.com/2015/06/24/analyses-of-the-oase-1-genome/

Awesome work, thanks!!

MitchellSince1893
07-04-2015, 01:03 PM
I've updated my list of Y-SNP calls for Oase 1. It should be clearer now.

From my revised post:

"The Y-SNP calls for Oase 1 show that he belonged to Y haplogroup K. He had positive calls for the IJK-defining SNPs PF3497 and PF3500, and for the K-defining SNP PF5495. Like Ust'-Ishim, Oase 1 had a positive call for the SNP Z4842 and a negative call for the SNP F650. These are two of the SNPs that define haplogroup X, which was created when the Y chromosome of an Indian Telugu was found to share seven mutations with haplogroup NO Y chromosomes. Like Ust'-Ishim, Oase 1 had only negative calls for NO-defining SNPs."

https://genetiker.wordpress.com/2015/06/24/analyses-of-the-oase-1-genome/


The Oase 1 sample is from a human mandible found in a cave in the Carpathian Mountains of southwestern Romania. The mandible is dated to 37,800–42,000 years BP.
FWIW

Yfull.com Ytree dates for Haplogroup K-M2335 (contains SNP Z4842): At http://www.yfull.com/tree/K-M2335/ put your cursor over "4 SNPs" to see Z4842

Formed 44000ybp (95% CI 48000<->40100 ybp)
TMCRA 41100 (95% CI 45200<->37000 ybp)

malleus
07-04-2015, 02:57 PM
LOL. Genetiker's new results still say that this guy belonged to K2a which includes NO and X. He may have belonged to the same lineage as Ust Ishim's. The rate of false positives is too high and it would be prudent to wait than jump to the conclusion. The only reliable conclusion is that he belonged to F, probably IJK. After that it is quite uncertain.

malleus
07-04-2015, 05:19 PM
FWIW

Yfull.com Ytree dates for Haplogroup K-M2335 (contains SNP Z4842): At http://www.yfull.com/tree/K-M2335/ put your cursor over "4 SNPs" to see Z4842

Formed 44000ybp (95% CI 48000<->40100 ybp)
TMCRA 41100 (95% CI 45200<->37000 ybp)

There are 7 SNPs altogether on that segment. Ust-Ishim has 2/5 positive and 2 had no calls if I remember correctly.
This guy had just 1 positive out of 2 good calls from that segment.
For Ust-Ishim one may very tentatively calculate (44000-41000)x2/5=1200 - > 42800 as the year his lineage diverged from the extant K2a line.

Assuming Genetiker's result is correct, 44000 roughly gives you the upper bound for Oase' age. 41000 does NOT give you the lower bound.
Oase's age is
the time of divergence(which will be between "formed" and TMRCA) - time to accumulate Oase's private SNPs
Genetiker does not provide much info for the latter. But I suspect that many "false positives" are actually recurrent private SNPs for Oase.

vettor
07-04-2015, 05:40 PM
LOL. Genetiker's new results still say that this guy belonged to K2a which includes NO and X. He may have belonged to the same lineage as Ust Ishim's. The rate of false positives is too high and it would be prudent to wait than jump to the conclusion. The only reliable conclusion is that he belonged to F, probably IJK. After that it is quite uncertain.

I did not see k2a where did you see this ?

was it not just k ?....................... and not K1, k2a or k2b

malleus
07-05-2015, 10:39 AM
I did not see k2a where did you see this ?

was it not just k ?....................... and not K1, k2a or k2b

Z4842 is a K2a marker. But he is negative at F650 which is another K2a marker. Ust Ishim was similar but had another positive and 4 other negatives.
If these are not errors this means that Oase and Ust Ishim man diverged early from the modern K2a line.

Power77
11-21-2016, 12:02 AM
I saw an Irish with haplogroup D* and also a Basque.

As much as I appreciate your posts, I must admit that I'm rather skeptical about that claim of yours. Are you 100% sure that these two individuals have no known "recent"(as in from "colonial" times [i.,e., from 16th to 19th century]) paternal East Asian and/or SSA ancestry? Did they "test" with 23andme or with FTDNA? As you can see, I'm very interested about the source where you "witnessed" these blokes;).

Arame
12-02-2016, 07:00 AM
A new person is added upstream of NO. He is from India.
https://yfull.com/tree/K-M2335/

Ebizur
12-02-2016, 08:34 AM
A new person is added upstream of NO. He is from India.
https://yfull.com/tree/K-M2335/I have noticed that K-M2335(xNO-M214) individual newly added to the YFull tree, too. His origin is stated to be Andhra Pradesh, which is the homeland of the Telugu people. The previously known example of K-M2335(xNO-M214) or "Haplogroup X" is a person of Telugu ethnic origin living in the UK, so the new example does not really expand the known distribution of this Y-DNA haplogroup at present. I hope additional members of K-M2335(xNO-M214) will have their Y-DNA tested and submit their data to YFull so that the TMRCA of extant members of the haplogroup may be estimated. (Personally, I am curious whether the patrilineal ancestors of extant Indian K-M2335(xNO-M214) have inhabited South Asia since the Pleistocene or whether they have migrated to India from elsewhere during the Holocene, e.g. from Southeast Asia or East Asia alongside the O-M95 of Austroasiatic-speaking peoples.)