PDA

View Full Version : My father's Y-DNA results



Alkaevli
09-13-2017, 04:19 PM
I got my father's Y12 results back today.

https://i.hizliresim.com/LOd9pa.png (https://hizliresim.com/LOd9pa)

Matches (50)

Page 1
https://i.hizliresim.com/ROEdEY.png (https://hizliresim.com/ROEdEY)

Page 2
https://i.hizliresim.com/mJqvq4.png (https://hizliresim.com/mJqvq4)


Matches Maps
https://i.hizliresim.com/0GbO4L.png (https://hizliresim.com/0GbO4L)


Y-DNA Haplotree
https://i.hizliresim.com/rJYPda.png (https://hizliresim.com/rJYPda)


Y-DNA Migration Maps
https://i.hizliresim.com/JOdP7n.png (https://hizliresim.com/JOdP7n)


Y-DNA Haplogroup Origins
https://i.hizliresim.com/0GbO8L.png (https://hizliresim.com/0GbO8L)


Y-DNA - Standard Y-STR Values
https://i.hizliresim.com/pGvrjn.png (https://hizliresim.com/pGvrjn)

Alkaevli
09-13-2017, 04:20 PM
My father's autosomal DNA is here: http://www.anthrogenica.com/showthread.php?11721-My-father-s-FTDNA-results-Sivas-Turkey

Afshar
09-13-2017, 04:37 PM
Do you have any tatar ancestry?
Member basmaci is also hg N, seems like there is a lot of N in Turkey

Alkaevli
09-13-2017, 05:35 PM
Do you have any tatar ancestry?
Member basmaci is also hg N, seems like there is a lot of N in Turkey
My father has no Tatar ancestry.

To be honest Cinnioğlu's famous study (Excavating Y-chromosome haplotype strata in Anatolia) was misleading, it was not based exclusively on ethnic Turks.

http://evolutsioon.ut.ee/publications/Cinnioglu2004.pdf
"A total of 359 samples were from blood banks, 61 from paternity clinics and 103 from staff and students enrolled at Istanbul University."

Haplogroup N is not rare at all judging from FTDNA Turkish projects, it is close to 10%.

RobertCasey
09-13-2017, 10:00 PM
Any 12 marker test will probably be able to assign you to a very ancient haplogroup - in your case haplogroup N. But the resolution of 12 markers, the average false positive rate is from 80 to 90 %. That yields a very inaccurate view of analysis performed based on FTDNA matches. At 25 markers, accuracy is dramatically increased to 40 to 60 % - but 50 % accuracy is still not very desirable. At 37 markers, accuracy improves to 60 to 80 % which is the first resolution that should be used for any detailed analysis. 67 markers does not improve accuracy that much but has many other extremely important factors that should be considered. 67 markers produces 75 to 90 % accuracy - so even at this level, it is somewhat random how much accuracy you will find.

The most important reason for 67 markers, there are other types of analysis that require this level of resolution. YSNP prediction at 1,500 to 2,500 years can yield 90 to 99 % accuracy. This is where 80 to 90 % of 67 markers testers can predict YSNPs in the 1,500 to 2,500 year range. At 37 markers, only around 5 to 10 percent of 37 markers can be predict YSNPs. Also, if you have 100 plus testers at 67 markers predicting a YSNP AND you have 20 to 50 YSNP branches below the predictable YSNP AND you have extensively tested YSNPs, charting becomes feasible at 60 to 95 % accuracy (depending the signature size used for prediction). Unfortunately, only around 10 to 20 % of current testers can really accurately chart since many factors are required for ideal conditions with current technology. Even with very favorable factors, only 10 to 80 % can not be charted driven by the amount of YSNP testing that is present.

Here are characteristics that allow robust charting: 1) being part of predictable single signature YSNP in the 1,500 to 2,500 year range that is very prolific in offspring is probably the most important factor. Regardless of geographical origins, the population that won battles vs. lost battles ended up with more offspring which increases the sample size of living testers. 2) The amount testing being done (percentage of population tested) is the second largest factor. But this requires large amounts of funding and there is very geographically bias in this factor. Testing of Asians is probably the most challenging due very small numbers of testers but very large populations. Surprisingly, Arabs are pretty well tested. However, higher YSTR resolution and YSNP testing is not as robust. Also, many Arab FTNDA projects are private in nature which limits analysis across a much broader crowd. Former Soviet countries and Scandinavian testers are now testing at about the same rates of western European which is very encouraging - but resolutions and YSNP testing is not as robust - but the groundwork for serious progress has now started.

Then there is just plain random statistical variation that messes things up for many testers. The more isolated you YSTRs are, the better results that you will have in be able to make accurate analysis. This is unfortunately is very random in nature across all geographies. My parents are both ends of the spectrum of random luck. My father's line is Irish (R-L226) which was associated with the first Irish king to unite/conquer the entire island of Ireland. We now have 570 testers at 67 markers, 90 Big Y tests, 90 SNP pack tests that are L226 only, etc. I am now able to accurately chart 81 % of the 570 testers.

My mother's line is just the opposite and barely survived to the present (even though I have published a family history that has 7,500 descendants). It is stuck at DF27 for years which is very old. I kept testing individual new branches under DF27 but they came back negative. However, it has been over a year since my last testing spree and that new elusive branch could have been discovered recently - but it will be very small in scope. This was the Brooks surname - so every fourth Brooks tester is new genetic cluster - living by a brook (stream) just does not help tie people together. Through YDNA research, I am now convinced my Brooks ancestor that was born in 1765 was probably adopted and was probably a Wade instead of Brooks. It is pretty sad that this has dramatically diminished my interest on this line. Being very less prolific in descendants and having an adoption in 1765. The only Brooks that are related with is the 1765 ancestor and one of his brothers - before that, I need to start over again with Wade research (not even very positive that this adoption is certain but is very likely).

Alkaevli
09-14-2017, 10:20 AM
Your post is helpful even for a noob like me, thanks.

Anabasis
09-14-2017, 06:54 PM
My father has no Tatar ancestry.

To be honest Cinnioğlu's famous study (Excavating Y-chromosome haplotype strata in Anatolia) was misleading, it was not based exclusively on ethnic Turks.

http://evolutsioon.ut.ee/publications/Cinnioglu2004.pdf
"A total of 359 samples were from blood banks, 61 from paternity clinics and 103 from staff and students enrolled at Istanbul University."

Haplogroup N is not rare at all judging from FTDNA Turkish projects, it is close to 10%.

well the aim.of the cinnioglus paper was not identify the genetic structure of ethnic turks but Turkis citizens. It was all about Turkey. Thus its objective but sample dsitrubition is still need to be discussed as far as there are low sample sizes in egean and medeterenian region.

Anabasis
09-14-2017, 07:00 PM
Additionaly your matches are mostly from central asian and russian region which show the recent ancestry from turkic migration. Its means there is low diversity in anatolia as far as most of the N samples from turkey match with each other. But it doesnt mean high frequency. On the other hand there are much more J and G samples who have lesser matches due to high diversity. Which means much more older shared ancestry in anatolia.

alhan
09-15-2017, 11:55 AM
Additionaly your matches are mostly from central asian and russian region which show the recent ancestry from turkic migration. Its means there is low diversity in anatolia as far as most of the N samples from turkey match with each other. But it doesnt mean high frequency. On the other hand there are much more J and G samples who have lesser matches due to high diversity. Which means much more older shared ancestry in anatolia.

I think Alkaevli is pretty lucky to have those matches on YDNA even with 12 markers.
My question would be; if his YDNA matches partly overlap with his autosomal matches.
If so I would read this as a sign of relatively recent migration.

In my case for YDNA I have just one match from Turkey who is irrelevant for the reasonable time frame.

I wouldn't expecting hundreds of matches of R1a from Turkey but in this way I feel pretty isolated in terms of YDNA data for the region.

Alkaevli
09-15-2017, 06:49 PM
well the aim.of the cinnioglus paper was not identify the genetic structure of ethnic turks but Turkis citizens. It was all about Turkey. Thus its objective but sample dsitrubition is still need to be discussed as far as there are low sample sizes in egean and medeterenian region.

That's true, but his paper is used as a reference for Turkish Y-DNA even today, he could have categorized the samples by ethnicity, like Grugni et al. 2012 (Iranian Y-DNA):

http://3.bp.blogspot.com/-9ItFg3ZDOCc/UAfrJKKY7aI/AAAAAAAAFC8/WQfxUO6_9Vw/w1200-h630-p-k-no-nu/journal.pone.0041252.t001.jpg


Most of the conclusions about Turkish Y-DNA is based on his paper, not to mention the fact that West Anatolia/Aegea (n=30/523), South Anatolia (n=33/523) and Western Black Sea (n=29/523) regions were underrepresented.

This pie chart from Oghuz Turks (https://www.familytreedna.com/groups/oghuz-turks/about) project is based on the results of 84 Anatolian Turks who have been tested, I think we can assume that hg N seems to be a significant Y-DNA marker among ethnic Turks although the sample size is currently insufficient to draw firm conclusions. My prediction is that the proportion of hg N will decrease and then be stabilized around 8-9% as the sample size grows larger.
https://i.hizliresim.com/JOE3mn.jpg

eolien
09-17-2017, 09:14 AM
You will not get an objective truth based on scientific methods from this type of data. 84 samples are not really randomly distributed. Many Tatars and others outside of turkey, at least 4 samples from the same place in Konya etc. What makes Tatars more turkish? They are modelled as 50/50 east/west euroasian. the same for people from turkmenistan and uzbekistan and uygurs. I agree that Q and N has some potential for a Turkic marker but much more study is needed.

Alkaevli
09-17-2017, 12:03 PM
You will not get an objective truth based on scientific methods from this type of data. 84 samples are not really randomly distributed. Many Tatars and others outside of turkey, at least 4 samples from the same place in Konya etc.

Hi eolien, let me clarify some of your points.

Non-Anatolian Turkic speakers (Tatars, Azerbaijanis, Balkan Turks etc.) are not included in that pie chart, all of those 84 individuals are Anatolian Turks. There are 146 individuals listed on the project's Y-DNA classic chart (some of them are private so if you are not a project member you can't see them) and 84 of them are Anatolian Turks (on whom the pie chart is based).

84 samples are actually randomly distributed, many Turkish provinces are represented despite the sample size being not large enough. You are right about Konya being overrepresented though, but as you can see each one of those 5 individuals from Konya has different haplogroups.
1 G-M201
1 I-L1242
1 J-ZS5891
1 N-VL67
1 R-PH200

Anyway, it is too early to draw conclusions and no one claims that the pie chart in the previous post reflects Turkish Y-DNA 100% accurately, but it will become more accurate as the sample size increases.

The point of my previous posts was clear, Cinnioğlu's methods were problematic. It's obvious that his aim was not to focus exclusively on ethnic Turks, nonetheless the regions should have been represented accurately.



What makes Tatars more turkish? They are modelled as 50/50 east/west euroasian. the same for people from turkmenistan and uzbekistan and uygurs. I agree that Q and N has some potential for a Turkic marker but much more study is needed.

Sorry, I did not get this part of your post. I don't remember anyone saying anything about Tatars being more Turkic.

By the way, Volga Tatars have 20% East Eurasian admixture on average (assuming the "Tatar" sample on gedmatch's spreadsheets represents Volga Tatars accurately), Turkmens and Uzbeks on the other hand seem to have 20-40% East Eurasian admixture depending on the region.

Afshar
09-17-2017, 08:04 PM
You will not get an objective truth based on scientific methods from this type of data. 84 samples are not really randomly distributed. Many Tatars and others outside of turkey, at least 4 samples from the same place in Konya etc. What makes Tatars more turkish? They are modelled as 50/50 east/west euroasian. the same for people from turkmenistan and uzbekistan and uygurs. I agree that Q and N has some potential for a Turkic marker but much more study is needed.

There is more "Turkic" than Q and N.

eolien
09-18-2017, 09:02 AM
There is more "Turkic" than Q and N.

such as R1a-Z93 or others?

Afshar
09-18-2017, 09:45 AM
such as R1a-Z93 or others?
Some clades of J2a

basmaci
11-11-2017, 09:00 AM
Do you have any tatar ancestry?
Member basmaci is also hg N, seems like there is a lot of N in Turkey
Yes, i see my name on the list:)
:D
https://media.giphy.com/media/d517Ozl6fqcb6/giphy.gif

Buhedyar
11-22-2017, 02:11 AM
My father has no Tatar ancestry.

To be honest Cinnioğlu's famous study (Excavating Y-chromosome haplotype strata in Anatolia) was misleading, it was not based exclusively on ethnic Turks.

http://evolutsioon.ut.ee/publications/Cinnioglu2004.pdf
"A total of 359 samples were from blood banks, 61 from paternity clinics and 103 from staff and students enrolled at Istanbul University."

Haplogroup N is not rare at all judging from FTDNA Turkish projects, it is close to 10%.

That study doesn't even deserve to be called a study.