PDA

View Full Version : New Population Finder Update, Finally?



Pages : [1] 2 3

Táltos
04-11-2014, 06:43 PM
Just received this email today from FTDNA, and nearly passed out. JK! :amen:

"Our new and vastly improved Population Finder is launching in just a few weeks! Soon, you'll be able to dive into fresh insights about your ethnic origins. You'll also be able to compare your ethnicty with your Family Finder matches!

If you want to share your ethnic origins with your matches, you don't need to take any action. You'll automatically be able to compare your ethnicity with your matches when the new Population Finder becomes available. This is the recommended option.

However, we do understand that sharing your ethnicity with your matches is your choice. Therefore, you may choose not to take part (opt-out). To opt-out, please follow the instructions below by April 30.*

Click this link, https://my.familytreedna.com/privacy-sharing.aspx.
If you are not logged in, do so.
Select the Do not share my ethnic breakdown with my matches radio button.
Click the Save button.
You may read more detailed instructions about this page in our Learning Center. You may also join our forums for discussion.

* You can change your privacy settings at any time. Thus, you may opt-out of or opt back into ethnic sharing at a later date if you change your mind."

lgmayka
04-11-2014, 09:58 PM
After all these years, FTDNA will stop labeling people of Polish ancestry as "Orcadian." :)

leonardo
04-11-2014, 10:05 PM
Just received this email today from FTDNA, and nearly passed out. JK! :amen:

"Our new and vastly improved Population Finder is launching in just a few weeks! Soon, you'll be able to dive into fresh insights about your ethnic origins. You'll also be able to compare your ethnicty with your Family Finder matches!

If you want to share your ethnic origins with your matches, you don't need to take any action. You'll automatically be able to compare your ethnicity with your matches when the new Population Finder becomes available. This is the recommended option.

However, we do understand that sharing your ethnicity with your matches is your choice. Therefore, you may choose not to take part (opt-out). To opt-out, please follow the instructions below by April 30.*

Click this link, https://my.familytreedna.com/privacy-sharing.aspx.
If you are not logged in, do so.
Select the Do not share my ethnic breakdown with my matches radio button.
Click the Save button.
You may read more detailed instructions about this page in our Learning Center. You may also join our forums for discussion.

* You can change your privacy settings at any time. Thus, you may opt-out of or opt back into ethnic sharing at a later date if you change your mind."
I too received this notice. I hope the stir over semargl won't cause people to overreact and make their newly offered information private. As many of you know, it is a challenge - as it is, to get matches share now.

evon
04-11-2014, 10:32 PM
After all these years, FTDNA will stop labeling people of Polish ancestry as "Orcadian." :)

Haha, and i thought it was bad enough when they did that to me :D

GTC
04-12-2014, 12:00 AM
After all these years, FTDNA will stop labeling people of Polish ancestry as "Orcadian." :)

Don't count your chickens ... yet ;)

rms2
04-12-2014, 01:37 PM
I was hoping FTDNA might have access to the People of the British Isles Project data for Population Finder. I wonder if that is the case. The whole "Orcadian" thing didn't bother me much, but I am kind of hoping to get some resolution out of that 4.67% I have that is currently listed as "Palestinian, Adygei, Bedouin, Bedouin South, Druze, Iranian, Jewish, Mozabite". (The other 95.33% is - yep, you guessed it - "Orcadian".)

Stephen1986
04-12-2014, 01:47 PM
I was hoping FTDNA might have access to the People of the British Isles Project data for Population Finder. I wonder if that is the case. The whole "Orcadian" thing didn't bother me much, but I am kind of hoping to get some resolution out of that 4.67% I have that is currently listed as "Palestinian, Adygei, Bedouin, Bedouin South, Druze, Iranian, Jewish, Mozabite". (The other 95.33% is - yep, you guessed it - "Orcadian".)

Yeah, my brother and myself have similar results to you. According to some analyses I have some significant continental ancestry, so hopefully the new PF will show that.

GTC
04-12-2014, 02:41 PM
I was hoping FTDNA might have access to the People of the British Isles Project data for Population Finder. I wonder if that is the case.

Given that the world is still awaiting publication of "the paper", I doubt that they will have released any of the data. However, one can only hope.

rms2
04-13-2014, 12:25 AM
Given that the world is still awaiting publication of "the paper", I doubt that they will have released any of the data. However, one can only hope.

That brings up something I have been wondering about for awhile: When will the POBI Project produce anything we can really sink our teeth into?

evon
04-13-2014, 02:27 AM
One thing FTDNA did get right that 23andme didnt, was the ANI and ASI separation in South Asia...looking forward to what they will do next, planning to test my mother also next month...

GTC
04-13-2014, 02:45 AM
That brings up something I have been wondering about for awhile: When will the POBI Project produce anything we can really sink our teeth into?

A very good question. The most recent suggestion I heard was "We might find out more in February when Bruce Winney will be talking about POBI at Who Do You Think You Are? Live", but that's come and gone and still no better information.

The longer this drags on, the more I'm inclined to think that the joint authors and/or their peer reviewers are at loggerheads over the conclusions. Perhaps they should just publish the data and let the rest of us arrive at the conclusions. :)

BMG
04-13-2014, 12:57 PM
One thing FTDNA did get right that 23andme didnt, was the ANI and ASI separation in South Asia...looking forward to what they will do next, planning to test my mother also next month...
what did they do ? i am not aware of that

evon
04-13-2014, 01:17 PM
what did they do ? i am not aware of that

They divide south Asian into ASI (Ancestral south India) and ANI (Ancestral north India), which is well a known north-south division within south asia..if you have tested with FTDNA (FamilyFinder), you can see this in Population finder..but i am hoping that an update will divide it even further :)

For info on ASI and ANI, see:

http://dienekes.blogspot.no/2013/08/major-admixture-in-india-took-place-42.html


A new paper on the topic of Indian population history has just appeared in the American Journal of Human Genetics. In previous work it was determined that Indians trace their ancestry to two major groups, Ancestral North Indians (ANI) (= West Eurasians of some kind), and Ancestral South Indians (ASI) (= distant relatives of Andaman Islanders, existing today only in admixed form). The new paper demonstrates that admixture between these two groups took place ~4.2-1.9 thousand years ago.

BMG
04-13-2014, 01:25 PM
They divide south Asian into ASI (Ancestral south India) and ANI (Ancestral north India), which is well a known north-south division within south asia..if you have tested with FTDNA (FamilyFinder), you can see this in Population finder..but i am hoping that an update will divide it even further :)

For info on ASI and ANI, see:

http://dienekes.blogspot.no/2013/08/major-admixture-in-india-took-place-42.html
Well i have tested with family finder and they are not making any such of divisions . I dont know from where you got this information .

evon
04-13-2014, 02:45 PM
Well i have tested with family finder and they are not making any such of divisions . I dont know from where you got this information .

It seems they have changed some of the links, i cant find the population description anymore, but in the main page it lists them as separated into: Southeast Indian and North Indian, which should be ASI and ANI....

BMG
04-13-2014, 03:44 PM
It seems they have changed some of the links, i cant find the population description anymore, but in the main page it lists them as separated into: Southeast Indian and North Indian, which should be ASI and ANI....

No that just means they have reference populations from north india and south east india which has nothing to do with ani or asi .

evon
04-13-2014, 10:05 PM
No that just means they have reference populations from north india and south east india which has nothing to do with ani or asi .

Its basically the same thing, since they would use populations from the north and south in finding components to create such division, but it dosnt matter anyways, as they will roll out a new Population finder by the end of April, so the old results are like spoiled meat, no need to keep it laying around...

soulblighter
04-13-2014, 11:16 PM
Its basically the same thing, since they would use populations from the north and south in finding components to create such division, but it dosnt matter anyways, as they will roll out a new Population finder by the end of April, so the old results are like spoiled meat, no need to keep it laying around...

BMG is right. It has nothing to do with ASI/ANI.
I am 100% North Indian on population finder, but I am sure I have substantial ASI (in fact all north Indians would).

BMG
04-14-2014, 03:46 AM
BMG is right. It has nothing to do with ASI/ANI.
I am 100% North Indian on population finder, but I am sure I have substantial ASI (in fact all north Indians would).
My results are as confusing
74.5% southasian
25.5% southasian(north indian ,southeast indian)

I didnt know what the first ref population mean but the second one includes north indian and southeast indian

Anyway my map includes whole of india,pakistan and afganistan .So they could be meaning my ancestry is could be from somewhere in there but that is too broad

soulblighter
04-14-2014, 01:16 PM
My results are as confusing
74.5% southasian
25.5% southasian(north indian ,southeast indian)

I didnt know what the first ref population mean but the second one includes north indian and southeast indian

Anyway my map includes whole of india,pakistan and afganistan .So they could be meaning my ancestry is could be from somewhere in there but that is too broad

That is strange indeed!
You have two different south Asian??
Can you post a screen shot?

BMG
04-14-2014, 03:20 PM
My Population finder results
1720

evon
04-14-2014, 03:28 PM
BMG is right. It has nothing to do with ASI/ANI.
I am 100% North Indian on population finder, but I am sure I have substantial ASI (in fact all north Indians would).


This is the division i was talking about:
http://i1010.photobucket.com/albums/af228/njal2010/See.jpg

I am sure you have ASI, but that dosnt meant the Old Population finder would show it, just wait for the new one. Again, i cant find the population sample overview, as the links have all changed to feature the new Population finder, which is why this debate is useless :P

BMG
04-14-2014, 03:36 PM
This is the division i was talking about:
http://i1010.photobucket.com/albums/af228/njal2010/See.jpg

I am sure you have ASI, but that dosnt meant the Old Population finder would show it, just wait for the new one. Again, i cant find the population sample overview, as the links have all changed to feature the new Population finder, which is why this debate is useless :P

Well Yeah!! . Eagerly waiting for the updated version .Hope it wont disappoint us .

Sein
04-14-2014, 07:39 PM
My current "Population Finder" results are highly unusual:

Middle East (Bedouin, Bedouin South, Mozabite, Palestinian)=96.79% ±0.60%
South Asian (Central Asian)=2.83% ±0.52%

They gave it a second try, and obtained the same results. I'm truly hoping that the update proves to be satisfactory. For what it's worth, Bennett added me to their % testing routine, so they did take this very seriously.

cyna
04-29-2014, 05:59 PM
Sure taking them a long time to get the update out. It has been almost 3 weeks. If you aren't less than 2 weeks away don't send out the notice!

Dr_McNinja
04-30-2014, 02:05 PM
They said it would be out after April 30th. I think they were aiming for May 1st but with the drama over their new Y-DNA Haplotree problems, it might be delayed.

FWIW, here's my old PF results, I think I may be one of the few for whom it worked:

http://i.imgur.com/qfte0Qf.png

Using these places as latitude/longitude coordinates (Lucknow for North India), it averaged me out to Pakistan, near my parents' village.

I'll probably get something weird/crazy in the new one :)

Táltos
04-30-2014, 02:50 PM
Yeah, I'm not optimistic about this at all. I also wonder if the Y Tree bungle will cause a hold up.

This was pointed out over on 23andme's thread about the update. https://www.familytreedna.com/learn/autosomal-ancestry/ethnic-origins/details-ethnicity-sharing-new-population-finder/

That is disappointing. They are only going to allow you to see the top 3 ethnicities that you share in common. Well good luck with that. I hope they are going to actually get it right. Believe me I have seen plenty of instances of where matches are sharing over at 23andme, and nobody is painted the same ethnicity where they match on the chromosome browser. I have seen this to be especially a big problem for my Mom's kit. In my kit they consistently get the matching and the Jewish correct. So they are really good at that over there. Unlike FTDNA who calls almost everyone Middle Eastern.

So how can I trust that FTDNA is going to get it right, when I feel 23andme is superior in this regard? Meaning even though 23andme is better, but they still get it wrong. And with FTDNA's track record...

Dr_McNinja
04-30-2014, 02:57 PM
That preview is interesting:

https://www.familytreedna.com/learn/wp-content/uploads/2014/04/popfinder.png

Looks like it shows Native American, North/East Asian (Japan, Korea) and some kind of Central European component. That last one is a big change.

Táltos
04-30-2014, 03:11 PM
That preview is interesting:

https://www.familytreedna.com/learn/wp-content/uploads/2014/04/popfinder.png

Looks like it shows Native American, North/East Asian (Japan, Korea) and some kind of Central European component. That last one is a big change.
A Central and Eastern European component would be excellent. I hope they have a thorough breakdown for all world populations. Well I'll keep my finger crossed for this update. But I know I shouldn't hold my breath for it!

cyna
04-30-2014, 03:22 PM
23andme utilizes fewer Ancestral SNPs than everyone else and they impute (I.e. guess) the rest (23andme only tests 500K SNPs and only half of those were ancestral). So with the worst coverage, I actually trust 23andme the least. I am just going to wait and see what FTDNA comes up with, with no set expectations as to quality.

cyna
04-30-2014, 03:23 PM
Is there a link to them saying after April 30?

Táltos
04-30-2014, 03:32 PM
23andme utilizes fewer Ancestral SNPs than everyone else and they impute (I.e. guess) the rest (23andme only tests 500K SNPs and only half of those were ancestral). So with the worst coverage, I actually trust 23andme the least. I am just going to wait and see what FTDNA comes up with, with no set expectations as to quality.
I know that this was a concern when they went to the v4 chip. I wish I could remember exactly what John O. of Gedmatch said about it over on 23andme's forums or find the thread. But he had indicated that what 23andme was doing with having less SNPs wasn't as big a deal. If I can find it I will post it.

Really all these calculators are personal preference, and what works so good for one will not work so good for another. Even though I'm on v4, 23andme did a pretty good breakdown. Much better than the current PF. Well I'll let you know what I think when my new results come in!

Táltos
04-30-2014, 03:33 PM
Is there a link to them saying after April 30?

They sent emails, that said if you want to opt out before the new update, to do it by April 30th.

Táltos
04-30-2014, 04:09 PM
23andme utilizes fewer Ancestral SNPs than everyone else and they impute (I.e. guess) the rest (23andme only tests 500K SNPs and only half of those were ancestral). So with the worst coverage, I actually trust 23andme the least. I am just going to wait and see what FTDNA comes up with, with no set expectations as to quality.
cyna I found the thread. See page 7, J Olson. He also has a comment on the last page to sum everything up again. Anyway his thoughts on the v4 made me feel better about it.

https://www.23andme.com/you/community/thread/25576/7/

Táltos
04-30-2014, 04:42 PM
If anybody is interested here is how many SNPs FTDNA uses for their current PF. https://www.familytreedna.com/learn/autosomal-ancestry/ethnic-origins/number-snps-population-finder-uses/

I wonder if this will change with the update?

evon
04-30-2014, 05:09 PM
Lets hope it will be released 1st may :) The new map looks interesting, but also very general, but there might be several layers to it, as with Ancestry Composition over at 23andme.

cyna
04-30-2014, 07:27 PM
Thanks Taltos. Yeh, I saw that article, and it won't be a big deal for most customers, who are largely one race or another. For someone highly admixed imputation is going to create a whole host of potential errors. Imputation is synonymous with guessing. If you are talking about someone mixed from mixed parents, then small segments are going to be thrown under the rug, and the admixture estimation is HIGHLY likely to be inaccurate. If you are 100% Euro, then imputation doesn't have that great an effect. If you have 1% Amerindian, as it appears I do, then that can get lost in the shuffle. If you have small segments of African, then 23andme is more likely to miss it and assign that chromosomal section as something else. I hear what was said, but logically, the less information you put in, the less reliable the results, by necessity. There is no way around that. And since 23andme looks to be taking a step back in the quality of their product (for economic reasons, which is fair), then I have stopped recommending them to people. They don't offer the health information and their genetic analysis (which was always an afterthought) now has less coverage than every other testing source. Sorry, but I don't have much confidence in their results.

Táltos
04-30-2014, 11:15 PM
Thanks Taltos. Yeh, I saw that article, and it won't be a big deal for most customers, who are largely one race or another. For someone highly admixed imputation is going to create a whole host of potential errors. Imputation is synonymous with guessing. If you are talking about someone mixed from mixed parents, then small segments are going to be thrown under the rug, and the admixture estimation is HIGHLY likely to be inaccurate. If you are 100% Euro, then imputation doesn't have that great an effect. If you have 1% Amerindian, as it appears I do, then that can get lost in the shuffle. If you have small segments of African, then 23andme is more likely to miss it and assign that chromosomal section as something else. I hear what was said, but logically, the less information you put in, the less reliable the results, by necessity. There is no way around that. And since 23andme looks to be taking a step back in the quality of their product (for economic reasons, which is fair), then I have stopped recommending them to people. They don't offer the health information and their genetic analysis (which was always an afterthought) now has less coverage than every other testing source. Sorry, but I don't have much confidence in their results.
Your welcome cyna. Sure the imputation had me concerned too. But as it turns out their guessing, actually guessed pretty good for me.

They were able to separate out the Jewish/Ashkenazi, Middle Eastern, Italian, and Balkan for me. And at least assign a large part of my AC to being Eastern European, even though they can not break down Eastern Europe much further.
Whereas FTDNA could not do this for me. Well I'll see how I feel after their update, LOL.

I also want to add, I hope that FTDNA starts to correctly assign people their Native American percentages with the update. I have seen them not give any NA to someone with known Huron aka Wyandot. Where 23andme found NA up to at least 1.6% for this person. I also have seen how this same kit on Gedmatch (uploaded with their FTDNA data) consistently scored 1-2% of Amerindian across calculators.

cyna
05-01-2014, 02:03 AM
It's funny, but everyone but 23andme and FTDNA, found 1-1.5% Amerindian for me. David from Eurogenes placed it in Central America. I am hoping Professor McDonald will respond to my submission and confirm.

Ahaddad
05-01-2014, 02:37 AM
The new family Finder are ready? Today is april 30...

rod
05-01-2014, 03:12 AM
Posted in the ISOGG Facebook group 30 April 2014 about 9 p.m. edt:
(About the new Population finder)

David Mittelman: I apologize everyone for providing the wrong info. We are rolling out in beta mode but it will be a little longer before everyone sees the release.

Táltos
05-01-2014, 03:50 AM
Posted in the ISOGG Facebook group 30 April 2014 about 9 p.m. edt:
David Mittelman: I apologize everyone for providing the wrong info. We are rolling out in beta mode but it will be a little longer before everyone sees the release.
(About the new Population finder)
Big sigh.......so what is a little longer. Another five years?

GTC
05-01-2014, 04:06 AM
Posted in the ISOGG Facebook group 30 April 2014 about 9 p.m. edt:
(About the new Population finder)

Thanks for that update.

The original PF is still tagged as "BETA" -- making it possibly the longest Beta release in IT history.

This time I'm hoping for a BETTER release.

cyna
05-01-2014, 12:28 PM
FTDNA is a complete joke. My theory is that like the Y-Tree they probably hit a snag with PF and as a face saving measure, they are claiming the rollout was just in BETA, but they forget to tell everyone.

Little bit
05-01-2014, 11:19 PM
The original PF is still tagged as "BETA" -- making it possibly the longest Beta release in IT history.

LOL...how did I never notice that before?!

http://i953.photobucket.com/albums/ae20/LB234/th_Screenshot2014-05-01at61354PM_zps0909581b.png (http://s953.photobucket.com/user/LB234/media/Screenshot2014-05-01at61354PM_zps0909581b.png.html)


Big sigh.......so what is a little longer. Another five years?

Quite possibly, yes?

icebreaker
05-01-2014, 11:38 PM
this is it

Sein
05-02-2014, 12:13 AM
That's actually pretty good for an ADMIXTURE-based test.

soulblighter
05-02-2014, 12:30 AM
this is it

looks great! Although looks like I am going to be 100% north Indian or 100% south Asian again...
I sometimes think it is a good thing that FDA screwed 23andme's health stuff, and so hopefully we will see some competition at play in the ancestry area, making both of them better!

Humanist
05-02-2014, 12:35 AM
This is actually less impressive than the previous Population Finder. According to this test I am 87% ~Turkish ("Anatolian Crossroads") and 13% Levantine ("Eastern Afroasiatic").

This is what happens when you do not have a sufficient population dataset.

Sein
05-02-2014, 12:40 AM
looks great! Although looks like I am going to be 100% north Indian or 100% south Asian again...
I sometimes think it is a good thing that FDA screwed 23andme's health stuff, and so hopefully we will see some competition at play in the ancestry area, making both of them better!

I'm betting that I'll be overwhelmingly "Eurasian Heartland" (it seems to peak in Afghanistan). But, at least I'm no longer going to be a Bedouin with a little Northeast Asian admixture thrown in. :biggrin1:

Humanist
05-02-2014, 12:59 AM
This is actually less impressive than the previous Population Finder. According to this test I am 87% ~Turkish ("Anatolian Crossroads") and 13% Levantine ("Eastern Afroasiatic").

This is what happens when you do not have a sufficient population dataset.

The southern Iraqi Mandaean most similar to me, under the new Population Finder:

Anatolian Crossroads: 83%
Eastern Afroasiatic: 15%
Central/South Asian: 2%

What do South Iraqi Mandaeans have to do with Turkey? Their language, way of life, etc. attest to a very long presence in what is today S Iraq. Obviously, the results are biased toward populations included in the dataset.

Táltos
05-02-2014, 01:57 AM
Well just noticed this was in my email.

E-mail not displaying correctly? View it in your browser.
DNA Tests | Login

myOrigins, our new version of Population Finder, is launching very soon! It will be available for FREE to everyone that has purchased a Family Finder test. Get ready to dive into your ethnic origins like never before!

With myOrigins, you'll be able compare your ethnicity with your Family Finder matches. If you want to share your ethnic origins with your matches, you don't need to take any action. You'll automatically be able to compare your ethnicity with your matches when myOrigins becomes available. This is the recommended option.

However, we do understand that sharing your ethnicity with your matches is your choice so we're sending you this reminder in case you want to not take part (opt-out). To opt-out, please follow the instructions below. *
Click this link.
If you are not logged in, do so.
Select the "Do not share my ethnic breakdown with my matches. This will not let me compare my ethnicity with my matches." radio button.
Click the Save button.
You can get more details about what will be shared here. You may also join our forums for discussion.

But when is soon for the rest of us???

Táltos
05-02-2014, 02:22 AM
I won't quote the person here completely, but someone who reports they are a project administrator said that he's "seen several results and most will be happy." (Southern Italians) This was over at 23andme.
He also indicated that, " They're tweaking browser & operating system issues now."---Which is kind of scary considering their IT nightmares.

Geez, this is kind of like waiting for the birth of a baby already!

Táltos
05-02-2014, 02:32 AM
Vinnie P. just told me over at 23andme it was ok to quote him about new PF here.


I've not heard anything about not updating the kits that were transferred - that would be ludicrous. Since I'm a project administrator, I've seen several results, and I think most people will be quite happy with the lay out as long as their analyses are what they're hoping for. One South Italian kit that was showing 100% ME, 91% Jewish, is now corrected to show both European and ME in expected amounts for that region, with about 6% Ashkenazi. They're tweaking browser & operating system issues now.

He also posted this.

Taltos, no, I'm not, but feel free to share. Actually, I've been able to see some limited updated results from a transfer from here, and I just saw an ISOGG forum posting which confirmed that transfers will definitely be included.

GTC
05-02-2014, 02:32 AM
I think access to myOrigins is currently restricted to Project Administrators.

Humanist
05-02-2014, 02:32 AM
I won't quote the person here completely, but someone who reports they are a project administrator said that he's "seen several results and most will be happy." (Southern Italians) This was over at 23andme.

I believe in the previous version of the Population Finder, some S Italians were appearing as "Jewish," correct?

EDIT: Never mind. Just saw your subsequent post.

Táltos
05-02-2014, 02:33 AM
I believe in the previous version of the Population Finder, some S Italians were appearing as "Jewish," correct?
Pretty much, or some other Middle Eastern.

Humanist
05-02-2014, 02:48 AM
One thing that the FTDNA tool has over the comparable 23andMe tool is that at least the former does not suggest that I am anything less than ~100% "Middle Eastern." Though, of course, this all depends on what one defines as "Middle Eastern."

http://i1178.photobucket.com/albums/x372/paulgiva78/passover/humanist_ethnic_makeup.png

BMG
05-02-2014, 02:58 AM
One thing that the FTDNA tool has over the comparable 23andMe tool is that at least the former does not suggest that I am anything less than ~100% "Middle Eastern." Though, of course, this all depends on what one defines as "Middle Eastern."

From where are u seeing the results . I am not seeing any ?

MitchellSince1893
05-02-2014, 03:29 AM
From where are u seeing the results . I am not seeing any ?

FTDNA Project admins can see them.

lgmayka
05-02-2014, 03:39 AM
FTDNA Project admins can see them.
Some can. I can't.

Ahaddad
05-02-2014, 04:01 AM
Some adm o ftdna can see the pf of me and my dad please? Kits 310205 and 298933

Humanist
05-02-2014, 04:08 AM
Some adm o ftdna can see the pf of me and my dad please? Kits 310205 and 298933

You are 56% Middle Eastern/44% European.

Your father is 98% Middle Eastern/2% Central/South Asian.

For further details regarding the breakdown, please refer to your FTDNA page when the update becomes accessible to all.

Dr_McNinja
05-02-2014, 04:18 AM
You are 56% Middle Eastern/44% European.

Your father is 98% Middle Eastern/2% Central/South Asian.

For further details regarding the breakdown, please refer to your FTDNA page when the update becomes accessible to all.Unrelated but I just noticed your Y-DNA hg is G1*. My brother in law is also G1* :) You're the first other person with that I've run into.

Ahaddad
05-02-2014, 04:21 AM
You are 56% Middle Eastern/44% European.

Your father is 98% Middle Eastern/2% Central/South Asian.

For further details regarding the breakdown, please refer to your FTDNA page when the update becomes accessible to all.

thanks you can show de % of populations? the ethnic

Dr_McNinja
05-02-2014, 04:28 AM
The southern Iraqi Mandaean most similar to me, under the new Population Finder:

Anatolian Crossroads: 83%
Eastern Afroasiatic: 15%
Central/South Asian: 2%

What do South Iraqi Mandaeans have to do with Turkey? Their language, way of life, etc. attest to a very long presence in what is today S Iraq. Obviously, the results are biased toward populations included in the dataset.What's Eastern Afroasiatic?

cyna
05-02-2014, 04:37 AM
Can anyone look and see what B9875 is?

razyn
05-02-2014, 05:02 AM
FTDNA Project admins can see them.

But not if they use a MacBook with Firefox as the browser... that is, I got a strangely truncated version of the map, that didn't show any of Europe (for a guy whose ancestry is shown as 100% European). It was roughly an equatorial slice, from Latin America on the left to east Africa or so on the right. Could not scroll on the map to change the view.

Switched to Safari for the browser, and then the map worked.

cyna
05-02-2014, 05:13 AM
razyn - Would you mind looking me up?

ZephyrousMandaru
05-02-2014, 05:47 AM
One thing that the FTDNA tool has over the comparable 23andMe tool is that at least the former does not suggest that I am anything less than ~100% "Middle Eastern." Though, of course, this all depends on what one defines as "Middle Eastern."

http://i1178.photobucket.com/albums/x372/paulgiva78/passover/humanist_ethnic_makeup.png

What a disappointment this is. I was looking forward to this new Population Finder too, what are they defining as "Anatolian Crossroads" anyway? What's with the vague population descriptions? These labels are worse than 23andMe's.

rod
05-02-2014, 06:13 AM
The old "Europe (Western European) Basque, French, Orcadian, Spanish" is now "European Coastal Plain - From the Bay of Biscay near Spain toward the Pripet Marshes of western Russia to the coastal plain of Northern Europe.":\

evon
05-02-2014, 09:38 AM
Examples :

http://www.norwaydna.no/family-finder/population-finder/

looks interesting :)

Dr_McNinja
05-02-2014, 09:39 AM
What a disappointment this is. I was looking forward to this new Population Finder too, what are they defining as "Anatolian Crossroads" anyway? What's with the vague population descriptions? These labels are worse than 23andMe's.I think they're purposely trying to use ancient terms (like that Eurogenes calculator which uses hunter gatherer, etc) so there's less controversy over people getting continents they didn't expect. Anatolian Crossroads probably just means Turkey/Caucasus but "Crossroads" signifies the region was a literal crossroads of many migrations of people. I wouldn't get any of that but my haplogroup for example, J-M241, probably was born around the "Anatolian Crossroads".

This would have fallen under 'Middle East' of the old Population Finder.

Dr_McNinja
05-02-2014, 09:42 AM
Examples :

http://www.norwaydna.no/family-finder/population-finder/

looks interesting :)So what would British/Irish people be getting in this? Coastal Plains?

Dr_McNinja
05-02-2014, 09:43 AM
I don't suppose one of the admins could look up which South Asian components are there in the new PF? Or is there a listing of all the possible components somewhere?

evon
05-02-2014, 09:45 AM
So what would British/Irish people be getting in this? Coastal Plains?

looks like it yes..i will ask a project mod to send me a screenshot of my results, as some of them can see the results already :)

rivergirl
05-02-2014, 10:31 AM
My mainly English/Irish ancestry with some Danish/Swedish and a bit of French show up as;
European coastal plain 70%, European Northlands 28% and North Mediterranean Basin 2%.
No European Coastal Islands eg UK, though European coastal plain map does take in the south coast of England.

My mum has;
European Northlands 43%, European coastal plains 29%, North Med Basin 17%, European coastal islands 10%.

A cousin with English, Scottish and German ancestry has;
European coastal islands 89% and European northlands 11%.
I thought this relative would have had more European Coastal Plains to account for the German ancestry.

We were all 100% Orcadian in the old version.

Another relative who was 94% Orcadian and 6% Middle Eastern is now;
98% European (Euopean coastal islands 60%, European northlands 19%, European coastal plains 19%)
and 2% Central/South Asian (Eurasian Heartland, which the map centres on Afghanistan).

rms2
05-02-2014, 10:54 AM
I had a look at the new "My Origins" remake of the old Population Finder this morning, and I like it. I didn't have time to look it over as thoroughly as I would like, but thus far I think it is a big improvement.

I am now "100% European": 53% European Coastal Plain, 43% European Coastal Isles, and 4% North Circumpolar.

soulblighter
05-02-2014, 10:54 AM
Got access to mine. I must say it is better than seeing a result that says 100% X ! but not much better than that.
1797

Under Central/South Asian, I am 58% "Indian Tectonic Plate" and 28% "Eurasian Heartland"
Under East Asian, I am all 8% "East Asian Coastal Islands" (note there was also a radiating spot from Madagascar in there but got clipped in my pic. I guess it is the Austonesian migrations.)
Under European, I am all 6% "Trans-Ural Peneplain"

Their description claims the following:

Indian Tectonic plate - Ancient South Asian hunter gatherers + Indo Aryans
Eurasian Heartland - Farmers who migrated through Iran and settled in Afghanistan/Pakistan/India
East Asian Coastal Islands - Tai/Kadai associated with Rice agriculture, and "adventurous" sailors.
Tras-Ural Peneplain: Associated with migration of nomads from Iran steppes (they seem to hint at WHG+ANE+EEF)

Dr_McNinja
05-02-2014, 11:00 AM
My mainly English/Irish ancestry with some Danish/Swedish and a bit of French show up as;
European coastal plain 70%, European Northlands 28% and North Mediterranean Basin 2%.
No European Coastal Islands eg UK, though European coastal plain map does take in the south coast of England.

My mum has;
European Northlands 43%, European coastal plains 29%, North Med Basin 17%, European coastal islands 10%.

A cousin with English, Scottish and German ancestry has;
European coastal islands 89% and European northlands 11%.
I thought this relative would have had more European Coastal Plains to account for the German ancestry.

We were all 100% Orcadian in the old version.

Another relative who was 94% Orcadian and 6% Middle Eastern is now;
98% European (Euopean coastal islands 60%, European northlands 19%, European coastal plains 19%)
and 2% Central/South Asian (Eurasian Heartland, which the map centres on Afghanistan).That's interesting. Now I'm really curious which populations go with these components. European coastal islands and North Mediterranean basin could be Atlantic (i.e, Cornwall, Southwest Scotland, Ireland) and Sardinian. Northlands is Scandinavian, coastal plains is probably Germanic. So your mom would be like 27% Atlantic-Mediterranean, 29% Germanic, 43% Northern Europe. Is the Irish ancestry on your mother's side recently and is the Danish/French on your father's side?

Dr_McNinja
05-02-2014, 11:07 AM
Got access to mine. I must say it is better than seeing a result that says 100% X ! but not much better than that.
1797

Under Central/South Asian, I am 58% "Indian Tectonic Plate" and 28% "Eurasian Heartland"
Under East Asian, I am all 8% "East Asian Coastal Islands"
Under European, I am all 6% "Trans-Ural Peneplain"LOL. I can't believe they named it Indian Tectonic Plate, they are really reaching far back into history. Perhaps Eurasian Heartland is related to Gedrosian?

rivergirl
05-02-2014, 11:25 AM
That's interesting. Now I'm really curious which populations go with these components. European coastal islands and North Mediterranean basin could be Atlantic (i.e, Cornwall, Southwest Scotland, Ireland) and Sardinian. Northlands is Scandinavian, coastal plains is probably Germanic. So your mom would be like 27% Atlantic-Mediterranean, 29% Germanic, 43% Northern Europe.
Is the Irish ancestry on your mother's side recently and is the Danish/French on your father's side?
My dads line is Irish mother from Co Laois and northern Ireland and English father, from East Anglia 1900s and Derby/Notts 1600-1840.

Mum has Irish from Co Clare and Galway 1830s, English from Essex/Herts and Devon 1820-50s, Danish 1870, Swedish 1800 and the French is from the late 1700s so shouldn't be so strong.

I therefore have more Irish and English than my mum, but it doesn't really come up as European coastal islands for me.
The North Med Basin map centres on Sicily and takes in north Africa, Greece, Yugoslavians, Italy, Switerland etc.
I'm looking at this via IPad, so can't see the descriptions for each category. They do mention they have 18 clusters.

soulblighter
05-02-2014, 11:28 AM
From where are u seeing the results . I am not seeing any ?
If you are part of a project, that admin may be able to see them. Most Syriac Christians seem to apparently show up as 91% south/central asian and 9% east Asian.


LOL. I can't believe they named it Indian Tectonic Plate, they are really reaching far back into history. Perhaps Eurasian Heartland is related to Gedrosian?

Yes that seems to be the case. I found the "Tectonic Plate" part funny too!

Dr_McNinja
05-02-2014, 11:37 AM
My dads line is Irish mother from Co Laois and northern Ireland and English father, from East Anglia 1900s and Derby/Notts 1600-1840.

Mum has Irish from Co Clare and Galway 1830s, English from Essex/Herts and Devon 1820-50s, Danish 1870, Swedish 1800 and the French is from the late 1700s so shouldn't be so strong.

I therefore have more Irish and English than my mum, but it doesn't really come up as European coastal islands for me.
The North Med Basin map centres on Sicily and takes in north Africa, Greece, Yugoslavians, Italy, Switerland etc.
I'm looking at this via IPad, so can't see the descriptions for each category. They do mention they have 18 clusters.What sticks out to me is Northern Ireland (which I presume would be the most English-like area of Ireland) and East Anglia on your father's side which is more Germanic (Coastal Plains). And Devon and Galway/Clare on your mother's side which would be more Coastal Islands. The French could contribute more to the Coastal Islands or Coastal Plains depending on where in France it was from so that's a bit of a wild card.

Dr_McNinja
05-02-2014, 11:40 AM
If you are part of a project, that admin may be able to see them. Most Syriac Christians seem to apparently show up as 91% south/central asian and 9% east Asian.



Yes that seems to be the case. I found the "Tectonic Plate" part funny too!It paints a picture of ancient Indians watching their continent smash into Asia. It looks like FTDNA is going for the opposite approach of 23andMe's Ancestry Composition which has tried to limit their groupings to modern times.

Very curious to see Sein and other Pashtun's results.

Dr_McNinja
05-02-2014, 11:44 AM
If you are part of a project, that admin may be able to see them. Most Syriac Christians seem to apparently show up as 91% south/central asian and 9% east Asian.



Yes that seems to be the case. I found the "Tectonic Plate" part funny too!Have we seen results of any Arabs yet? I wonder if they're using a Geno 2.0-like Southwest Asian component which combines Bedouin and Gedrosian.

rms2
05-02-2014, 11:50 AM
I'll try to take a screenshot when I get home later, but, although the European Coastal Plain (my biggest component, at 53%) blurb says it goes as far east as the Pripet Marshes of western Russian, the shading for it on my map extends only as far east as Austria and perhaps the southwest corner of the Czech Republic. The shading covers only westernmost Germany and is concentrated mainly on France, northern Italy, the Netherlands and on up into the southern British Isles. The version of European Coastal Plain that appears on my "myOrigins" map is very much western.

Otherwise, I got 43% European Coastal Isles, which appears on the map as the British Isles and the NW European continent, and 4% North Circumpolar, which is mainly Finland, far northern Sweden and Norway, and Karelia in Russia.

I like this new myOrigins thing very much.

rivergirl
05-02-2014, 11:54 AM
I found this for one of my project members.
East Asian Coastal Islands, map centred on Vietnam, Laos, southern China
Asian northeast, map centred on Japan

Anglecynn
05-02-2014, 12:17 PM
How do you access this new version? Mine still only has the old version.

Dr_McNinja
05-02-2014, 12:19 PM
Not sure but if anyone here is a project admin of one of the J haplogroup projects or the Mideast/Indian/etc groups, can you check mine out? Kit B6225

Hok
05-02-2014, 01:01 PM
Will people that ordered an autosomal dna transfer receive the new update aswell?

rivergirl
05-02-2014, 01:10 PM
LOL. I can't believe they named it Indian Tectonic Plate, they are really reaching far back into history. Perhaps Eurasian Heartland is related to Gedrosian?

Cut and Paste from FTDNA cluster description; (I started a thread with the cluster descriptions I found in my projects, maybe others can post the others.)

Eurasian Heartland
Eurasian Heartland can be found across a large band in the center of the Eurasian continent. It travels from the north of Europe south toward the narrow base of the Indian subcontinent. Within South Asia, it is the partner to the Indian Tectonic cluster. Across the center of Eurasia, it highlights the ancient influence of the Iranian nomads, Scythians, Sarmatians, and Cimmerians.
As early farmers moved west out of the hills of the Levant nearly 10,000 years ago, they mostly went into Europe and North Africa. The ancestors of the Eurasian Heartland, however, cautiously explored their possibilities to the east. They came into the light of history as the Persians, Sogdians, and Afghans, and have always had a hand on the turning points of history between the west and the east.
With the movement of millions from the Indian subcontinent, Eurasian Heartland has appeared in the Pacific, the New World, and Southeast Asia. Millions of displaced Afghans have also brought it west into Iran, and reinforced it in Pakistan.

soulblighter
05-02-2014, 01:31 PM
It paints a picture of ancient Indians watching their continent smash into Asia. It looks like FTDNA is going for the opposite approach of 23andMe's Ancestry Composition which has tried to limit their groupings to modern times.

Very curious to see Sein and other Pashtun's results.

It says the "Indian Tectonic" Population includes ancient hunter gatherers, and later migrants including Indo-Aryan migrations.

AJL
05-02-2014, 02:07 PM
So what would British/Irish people be getting in this? Coastal Plains?

For me, I am half "European Coastal Islands." {EDIT} But I think they're using a kind of smart matching, by comparing you to your matches. If so this makes a lot more sense because my second cousin is 100% European Coastal Islands.

Strangely I get 7% European Northlands (roughly Norway) rather than France/Germany, so I don't think the "Orkneys problem" has been entirely fixed. Looks like FTDNA now has the same situation Ancestry had when they first tried this.

The Jewish part is "Diaspora Jewish" which make more sense than either the old "Middle Eastern" or 23andme's "European : Ashkenazi."

Dr_McNinja
05-02-2014, 02:15 PM
Soulblighter looked up my results:


Central/South Asian
- Indian Tectonic=38%
- Eurasian Heartland=49%

European
- North Circumpolar 6%
- Trans-Ural Peneplain 3%

East Asian
- East Asian Coastal Islands 5%My original PF results: http://i.imgur.com/qfte0Qf.png

The original results were a lot more relevant but this isn't too bad. It's better for me than 23andMe and Geno 2.0 (just because of the Eurasia Heartland component) and not as bad as the old one was for some other people so I guess it's a better all around fit. It has some correlations to Geno 2.0 which gave me 50% Southwest Asian, 4% NE-Asian+Native American, and 11% Northern European. But those are probably superficial because of the weird Mediterranean component in Geno 2.0.

The Indian Tectonic Plate component is interesting, it's a combination of ANI and ASI, so some East Asian is pushed out. There's no real way to correlate that to existing admixture components that I can tell at the moment. I would love to see them reveal information on it, especially the reference populations and how it relates to those we've used in admixture. Could shed some light on the ANI-ASI relationship. As it stands I'm not sure how it relates to admixture results. Eurasian Heartland would incorporate Caucasian and Gedrosian (where I'd usually get 35%+10% separately or 25-30% as a joint component) and maybe a little bit of Western European (my Orcadian is gone and now Western Europeans are instead registering a little Eurasian Heartland, so that means there was a strong Gedrosian affinity for that mysterious Orcadian) but that still doesn't add up to 49% plus however much Indo-Aryan is in Indian Tectonic. So doing some math, 9%+49% = 58% West Eurasian. If I'm 80% West Eurasian, then 22% of my 38% Indian Tectonic is West Eurasian and roughly 21% total East Eurasian is left over (of which, some might be Arctic/Siberian/Native American/etc and West Eurasian in ancestry so ~17% ASI). Of course that doesn't help because what we'd like to know is the total West Eurasian amount and go from there!

I like the Orcadian/Gedrosian flip since it likely came into Europeans from Gedrosia and not into Gedrosia from Europe. Going by their explanation for Eurasian Heartland, they believe it was from those famous Sarmatians/Scythians or a similar population. I think most are leaning to that conclusion.

It looks like they tried to remove Caucasus as an ancestral component for populations not from the immediate vicinity. I guess it was pulling a lot of people to their old Middle Eastern category. So a lot of these new components incorporate Caucasian elements within them.

Dr_McNinja
05-02-2014, 02:29 PM
For me, I am half "European Coastal Islands." {EDIT} But I think they're using a kind of smart matching, by comparing you to your matches. If so this makes a lot more sense because my second cousin is 100% European Coastal Islands.

Strangely I get 7% European Northlands (roughly Norway) rather than France/Germany, so I don't think the "Orkneys problem" has been entirely fixed. Looks like FTDNA now has the same situation Ancestry had when they first tried this.

The Jewish part is "Diaspora Jewish" which make more sense than either the old "Middle Eastern" or 23andme's "European : Ashkenazi."I'm not sure if that's a fixable problem. Orcadian ancestry includes Germanic, Celtic/Atlantic, as well as Norse, so some of your ancestry is so mixed it will always pull towards Norway. Another way of looking at it is that it's pulling to "North Sea" (seen in populations from Kent, England) which represents Northwestern France, Netherlands, Southeastern England, and Southwestern Norway. Being that there is no North Sea component here, it's split up among Coastal Plains, Coastal Isles, and Northlands. Some of your previous Orcadian is going into South/Central Asian for some Europeans too. They basically got rid of their problematic components (Mideastern and Orcadian) since it would mostly split up correctly with lesser amounts of confounding results (in this case people getting a few % Northlands is better than getting 100% Orcadian).

Anglecynn
05-02-2014, 02:32 PM
I don't know if this will work, but if people could look up other's results, could somebody look up mine? Kit number: B1913

soulblighter
05-02-2014, 02:55 PM
Soulblighter looked up my results:

My original PF results: http://i.imgur.com/qfte0Qf.png

The original results were a lot more relevant but this isn't too bad. It's better for me than 23andMe and Geno 2.0 (just because of the Eurasia Heartland component) and not as bad as the old one was for some other people so I guess it's a better all around fit. It has some correlations to Geno 2.0 which gave me 50% Southwest Asian, 4% NE-Asian+Native American, and 11% Northern European. But those are probably superficial because of the weird Mediterranean component in Geno 2.0.

The Indian Tectonic Plate component is interesting, it's a combination of ANI and ASI, so some East Asian is pushed out. There's no real way to correlate that to existing admixture components that I can tell at the moment. I would love to see them reveal information on it, especially the reference populations and how it relates to those we've used in admixture. Could shed some light on the ANI-ASI relationship. As it stands I'm not sure how it relates to admixture results. Eurasian Heartland would incorporate Caucasian and Gedrosian (where I'd usually get 35%+10% separately or 25-30% as a joint component) and maybe a little bit of Western European (my Orcadian is gone and now Western Europeans are instead registering a little Eurasian Heartland, so that means there was a strong Gedrosian affinity for that mysterious Orcadian) but that still doesn't add up to 49% plus however much Indo-Aryan is in Indian Tectonic. So doing some math, 9%+49% = 58% West Eurasian. If I'm 80% West Eurasian, then 22% of my 38% Indian Tectonic is West Eurasian and roughly 21% total East Eurasian is left over (of which, some might be Arctic/Siberian/Native American/etc and West Eurasian in ancestry so ~17% ASI). Of course that doesn't help because what we'd like to know is the total West Eurasian amount and go from there!

I like the Orcadian/Gedrosian flip since it likely came into Europeans from Gedrosia and not into Gedrosia from Europe. Going by their explanation for Eurasian Heartland, they believe it was from those famous Sarmatians/Scythians or a similar population. I think most are leaning to that conclusion.

It looks like they tried to remove Caucasus as an ancestral component for populations not from the immediate vicinity. I guess it was pulling a lot of people to their old Middle Eastern category. So a lot of these new components incorporate Caucasian elements within them.


I find the East Asian Coastal Plains interesting. It seems to exist in most people from India. If it doesn't have ASI as a part of it, then maybe pre-ASI?

ZephyrousMandaru
05-02-2014, 03:10 PM
I think they're purposely trying to use ancient terms (like that Eurogenes calculator which uses hunter gatherer, etc) so there's less controversy over people getting continents they didn't expect. Anatolian Crossroads probably just means Turkey/Caucasus but "Crossroads" signifies the region was a literal crossroads of many migrations of people. I wouldn't get any of that but my haplogroup for example, J-M241, probably was born around the "Anatolian Crossroads".

This would have fallen under 'Middle East' of the old Population Finder.

Yeah, but they're still using linguistic terms to denote populations such "Eastern Afroasiatic", which is just ridiculous. I don't see how using ordinary geographical terms effects their neutrality. I think these new labels are just going to cause more confusion with people, especially if they're unaware just how frivolous these labels are. The modality of the "Anatolian Crossroads" component doesn't imply an origin there, so it'd be inaccurate to simply refer to it as if it were an exclusively "Anatolian/Caucasus" component.

Caucasians are a highly genetically drifted population, which is the reason why it peaks here.

AJL
05-02-2014, 03:23 PM
I'm not sure if that's a fixable problem. Orcadian ancestry includes Germanic, Celtic/Atlantic, as well as Norse, so some of your ancestry is so mixed it will always pull towards Norway. Another way of looking at it is that it's pulling to "North Sea" (seen in populations from Kent, England) which represents Northwestern France, Netherlands, Southeastern England, and Southwestern Norway. Being that there is no North Sea component here, it's split up among Coastal Plains, Coastal Isles, and Northlands. Some of your previous Orcadian is going into South/Central Asian for some Europeans too. They basically got rid of their problematic components (Mideastern and Orcadian) since it would mostly split up correctly with lesser amounts of confounding results (in this case people getting a few % Northlands is better than getting 100% Orcadian).

That's true. In getting rid of the bogus Middle East, that's a definite improvement.

Humanist
05-02-2014, 03:26 PM
Yeah, but they're still using linguistic terms to denote populations such "Eastern Afroasiatic", which is just ridiculous. I don't see how using ordinary geographical terms effects their neutrality. I think these new labels are just going to cause more confusion with people, especially if they're unaware just how frivolous these labels are. The modality of the "Anatolian Crossroads" component doesn't imply an origin there, so it'd be inaccurate to simply refer to it as if it were an exclusively "Anatolian/Caucasus" component.

This is the most pertinent part of the "Anatolian Crossroads" description (please see below). Again, I must ask, how does this square with the fact that you have populations like S Iraqi Mandaeans also showing up as predominantly "Anatolian Crossroads." So, Druze, Assyrians, Iraqi Mandaeans, and presumably Eastern Mizrahim are all predominantly "Anatolian Crossroads?" How does that reconcile with reality? Iraqi Mandaeans spoke a language related to the long extinct Southeastern Aramaic (Central/South Iraq) dialect of the Babylonian Talmud.


http://i1178.photobucket.com/albums/x372/paulgiva78/passover/anatolian_crossroads.jpg

RyGuy
05-02-2014, 03:39 PM
Humanist, or anyone else with access, could you tell me the results for my father and I? We are 318482 and B7682 respectively. Many thanks!

ADW_1981
05-02-2014, 03:45 PM
This is the most pertinent part of the "Anatolian Crossroads" description (please see below). Again, I must ask, how does this square with the fact that you have populations like S Iraqi Mandaeans also showing up as predominantly "Anatolian Crossroads." So, Druze, Assyrians, Iraqi Mandaeans, and presumably Eastern Mizrahim are all predominantly "Anatolian Crossroads?" How does that reconcile with reality? Iraqi Mandaeans spoke a language related to the long extinct Southeastern Aramaic (Central/South Iraq) dialect of the Babylonian Talmud.

They should have labeled Eastern AfroAsiatic as just Levant, or East Mediterranean. They probably should have had a Mesopotamian component in addition to the current Anatolian one. That would have made things a little cleaner.

seferhabahir
05-02-2014, 03:47 PM
This is the most pertinent part of the "Anatolian Crossroads" description (please see below). Again, I must ask, how does this square with the fact that you have populations like S Iraqi Mandaeans also showing up as predominantly "Anatolian Crossroads." So, Druze, Assyrians, Iraqi Mandaeans, and presumably Eastern Mizrahim are all predominantly "Anatolian Crossroads?" How does that reconcile with reality? Iraqi Mandaeans spoke a language related to the long extinct Southeastern Aramaic (Central/South Iraq) dialect of the Babylonian Talmud.

Humanist,

Eastern Mizrahim are showing up as about 70% Anatolian Crossroads and 30% Eastern Afro-Asiatic

seferhabahir
05-02-2014, 03:55 PM
That's true. In getting rid of the bogus Middle East, that's a definite improvement.

Yes, I have to agree. I now show up as 99% Jewish Diaspora and 1% European (probably North Mediterranean Basin, although I can't actually see that breakdown yet) which is much more realistic than the old 100% Middle Eastern categorization. My autosomal affinities usually align with Ashkenazi, Sicilian, Greek, Southern Italian.

ZephyrousMandaru
05-02-2014, 04:35 PM
This is the most pertinent part of the "Anatolian Crossroads" description (please see below). Again, I must ask, how does this square with the fact that you have populations like S Iraqi Mandaeans also showing up as predominantly "Anatolian Crossroads." So, Druze, Assyrians, Iraqi Mandaeans, and presumably Eastern Mizrahim are all predominantly "Anatolian Crossroads?" How does that reconcile with reality? Iraqi Mandaeans spoke a language related to the long extinct Southeastern Aramaic (Central/South Iraq) dialect of the Babylonian Talmud.

Right, more than likely, they're mislabeling this component. If it's a principal component in the Druze, Iraqi Mandaeans, Assyrians and Eastern Mizrahim Jews. All populations which did not have their ethnogenesis in Anatolia, than something's is wrong here. Anatolian Crossroads should be renamed to Irano-Mesopotamian/Anatolian.

I think FTDNA is making the mistake of inferring an ancestral component's origin from where it peaks in modern populations. I think this "Anatolian Crossroads" component probably originated in the Fertile Crescent region, and became modal in the Caucasus due to genetic drift caused by isolation and inbreeding.

So it'd be incorrect to say that populations rich in this "Anatolian Crossroads" component, are native to the region. It very well could be the other way around.

Hanna
05-03-2014, 06:42 AM
My grandmother's population Finder update:

https://fbcdn-sphotos-f-a.akamaihd.net/hphotos-ak-ash3/t1.0-9/10252179_289823034526709_4510544170356792197_n.jpg

My grandfather's Population Finder update:

https://fbcdn-sphotos-b-a.akamaihd.net/hphotos-ak-prn2/t1.0-9/10152663_289822437860102_6879330329434202957_n.jpg

I assume the Central/South Asian is equal to the Gedrosian component for my grandfather.

Alexandre
05-04-2014, 12:48 AM
Hi there everyone. I would just like to know if someone could get the new myOrigins results for me please? I am really curious about it and unfortunatelly i can't see mine yet.

My kit number is: 322161

Thank you very much.

AJL
05-04-2014, 12:56 AM
Hi there everyone. I would just like to know if someone could get the new myOrigins results for me please? I am really curious about it and unfortunatelly i can't see mine yet.

My kit number is: 322161

Thank you very much.

What projects are you a member of?

Alexandre
05-04-2014, 01:09 AM
I'm in project Brasil. Just that one. If you want i can join another one so you can see my results. Thanks for replying.

Sein
05-04-2014, 01:09 AM
If anyone could possibly see my results, my kit number is H5678. I'm a member of the "R1a & all subclades" project, and the "R1a" project. I'd truly appreciate this.

shazou
05-04-2014, 01:17 AM
edit: nevermind i already got a preview from an admin by email

AJL
05-04-2014, 01:21 AM
If anyone could possibly see my results, my kit number is B4819. I'm a member of the "R1a & all subclades" project, and the "R1a" project. I'd truly appreciate this.

Sorry, I can't see your results. They are showing for some project members, but not you. :\

Sein
05-04-2014, 01:25 AM
Sorry, I can't see your results. They are showing for some project members, but not you. :\

But thank you very much for checking. :)

I suppose I just have to wait and see.

Dr_McNinja
05-04-2014, 01:45 AM
^ If you had a transfer from 23andMe they won't show up yet.

Sein
05-04-2014, 01:50 AM
^ If you had a transfer from 23andMe they won't show up yet.

Ah, that explains it.

Dr_McNinja
05-04-2014, 02:04 AM
Averaging up the latitude/longitude coordinates for my results puts me about 350-400km to the west of the old Population Finder (which put me in Punjab near my parents' village), but still just barely within the Pakistani border near Afghanistan. That's pretty close all things considered (that it could be off by so much more). I wonder if other Punjabis using this are pulled even further off to the west. On the other hand it's surprising that it would be so close to where we are today considering these populations are based off a time frame from probably 4,000 to 10,000 years ago.

Táltos
05-04-2014, 03:10 AM
That's true. In getting rid of the bogus Middle East, that's a definite improvement.
Okay so I was able to get a sneak peak at my double secret BETA results.

Here is my old one.

Western Europe-French, Orcadian-85.79% MOE=+/-5.69%
Middle East-Palestinian, Iranian, Jewish, Adygei, Bedouin, Bedouin South, Druze-14.21% MOE=+/-5.69%

And the Update?
European-83%
Middle East-12%
Jewish Diaspora-5%

The only huge improvements I can see are:
1. They can break down my Jewish from the rest of my Middle Eastern
2. Though I didn't have the breakdown of clusters sent to me, I can see on the heat map the Trans Ural region is the darkest red, and makes more sense than Orcadian. Though it looks like I have all areas of Europe colored in except Southern Italian, the Balkans, and Spain. Go figure.

So I guess I can infer this Middle Eastern result is ancient? It's bogus still? On 23andme I do get a small percentage of Middle Eastern, and surprisingly almost all of it is from my father's side. Only 0.1 Middle Eastern on that site is attributed to my Mom. Turkey and areas around it are darkest green on my map so I guess I'll get the Anatolian Crossroads. In lighter green I get Morocco, Tunisia, and Algeria.

Oh I don't know already!

rod
05-04-2014, 03:53 AM
Population Finder
Europe (Western European) Basque, French, Orcadian, Spanish 97.67% ±8.07%

myOrigins
European Coastal Plain 85%
European Coastal Islands 11%
European Northlands 5%
Total European 101%

Telfermagne
05-04-2014, 04:26 AM
If anyone could possibly see my results, my kit number is B4819. I'm a member of the "R1a & all subclades" project, and the "R1a" project. I'd truly appreciate this.

I'm also interested in a sneak peak :p 160082 cheers!

Telfermagne
05-04-2014, 04:50 AM
I'm also interested in a sneak peak :p 160082 cheers!
NVM forgot that I was an admin for a project that I started but haven't finished up (Mountain District of Middle Tennessee). Here's my new PF results:

100% European Coastal Plain

European Coastal Plain
The European Coastal Plain combines nearly all of the threads of European genetic history into one. This cluster goes from the Bay of Biscay near Spain, toward the Pripet Marshes of western Russia, to the coastal plain of Northern Europe. The hunter-gatherer, farmer, and intruder from the steppes were forged together as one people. The French and the German were created by the intersection between the civilized and the barbarian during antiquity. With this diverse ancestry across the uniform plain, a relatively unified cluster was born.
European Coastal Plain represents the diverse groups brought together over the past 5,000 years, as Germans, Celts, and Slavs have moved in with their cattle, and the Romans brought their mills and cities. This cluster is common among many populations with Northern European heritage. Germanic migrations after the fall of Rome guaranteed its presence in the south. The people on the European Coastal Plain are at the heart of recent history. Being the engines behind the Great Powers of the age, they became the dominant actors in colonization of the world.

Táltos
05-04-2014, 04:55 AM
Population Finder
Europe (Western European) Basque, French, Orcadian, Spanish 97.67% ±8.07%

myOrigins
European Coastal Plain 85%
European Coastal Islands 11%
European Northlands 5%
Total European 101%
I think both our updates are a rehash of the old.

mkdexter
05-04-2014, 05:04 AM
For those using their GAP project - trying to see 23andMe kit myOrigins and can't.

1. View the kit from GAP just like any other kit.
2. Replace the web address "https://my.familytreedna.com/default.aspx?ekit=xxxxxxxxxxxxxxxxxxxxxxx" with "https://my.familytreedna.com/my-origins/?ekit=xxxxxxxxxxxxxxxxxxxxxxx" and make sure the same code number is still listed at the end of the web address. so.. default.aspx is to be replaced with my-origins/
3. This will bring up the transfer kits myOrigins page.

Telfermagne
05-04-2014, 05:08 AM
I'm curious as to why My Origins has not picked up the 0.1% West-African and 0.5% Ashkenazi that 23andMe picked up (both remain at all 3 thresholds so it seems to be real admixture). Otherwise, "My Origins" seems okay - the 100% European Coastal Plain result is consistent with the "location" I get when I average and coordinate my past 23andMe results & Gedmatch results.

Volat
05-04-2014, 08:14 AM
Population Finder @ FTDNA is inaccurate for me ancestry. 74% Orcadian, 25% Russian at almost no margin of error, which is not even close.

Yggdrasil
05-04-2014, 08:41 AM
I'm curious as to why My Origins has not picked up the 0.1% West-African and 0.5% Ashkenazi that 23andMe picked up (both remain at all 3 thresholds so it seems to be real admixture). Otherwise, "My Origins" seems okay - the 100% European Coastal Plain result is consistent with the "location" I get when I average and coordinate my past 23andMe results & Gedmatch results.

I read they put in a threshold of 1% in My Origins.

evon
05-04-2014, 11:07 AM
Population Finder
Europe (Western European) Basque, French, Orcadian, Spanish 97.67% ±8.07%

myOrigins
European Coastal Plain 85%
European Coastal Islands 11%
European Northlands 5%
Total European 101%

101%, wow, über european ;)

rms2
05-04-2014, 12:01 PM
The old Population Finder has me at about 96% European (all Orcadian) and about 4% "Middle Eastern". MyOrigins has me at 100% European: 53% European Coastal Plain, 43% European Coastal Isles, and 4% North Circumpolar.

I am somewhat surprised at that result and would have expected the Coastal Plain and Isles percentages to be reversed, at least. I would think my Isles percentage should be much higher, since almost all my ancestry that I know about is British Isles, but then again, there are a number of big holes in my family tree, and my dad is from Louisiana, so I suspect he may have a number of French lines that I don't yet know about that are driving up my European Coastal Plain score. I already have at least one French and one Dutch line on my dad's side that I do know about.

Anyway, myOrigins looks like a big improvement over Population Finder.

Dr_McNinja
05-04-2014, 12:37 PM
There's lots of "Coastal Plains" (i.e, Germanic) influence in England, especially in the Southeastern parts. I think some people from Kent would turn up overwhelmingly Coastal Plains.

rms2
05-04-2014, 12:41 PM
There's lots of "Coastal Plains" (i.e, Germanic) influence in England, especially in the Southeastern parts. I think some people from Kent would turn up overwhelmingly Coastal Plains.

The map shading for European Coastal Plain does extend into southern England, so that could be a factor. You're right.

AJL
05-04-2014, 03:34 PM
So I guess I can infer this Middle Eastern result is ancient? It's bogus still? On 23andme I do get a small percentage of Middle Eastern, and surprisingly almost all of it is from my father's side. Only 0.1 Middle Eastern on that site is attributed to my Mom. Turkey and areas around it are darkest green on my map so I guess I'll get the Anatolian Crossroads. In lighter green I get Morocco, Tunisia, and Algeria.

Oh I don't know already!

Tough to say. What I meant, primarily, is the bogus Middle Eastern turning up in Island people simply because they deviated eastwards relative to the Orkneys cluster. Orcadians are quite drifted so it was not a good proxy population for all English, Irish, Scots, etc.

rod
05-04-2014, 03:43 PM
After all the hype and years of waiting, people are going to be so disappointed. Maps and cute names and descriptions aren't going to cover the fact that they haven't really changed anything.

Humanist
05-04-2014, 03:50 PM
After all the hype and years of waiting, people are going to be so disappointed. Maps and cute names and descriptions aren't going to cover the fact that they haven't really changed anything.

I do not like the fact that they mention ancient populations in their component descriptions. As far as I know, there are no tested DNA remains of Hittites, Phrygians, Assyrians, Persians, etc. dating to periods further back than the present. I can see if they were available, then such descriptions would potentially be relevant. Otherwise, it just strikes me as a bit gimmicky, and perhaps misleading.

Táltos
05-04-2014, 03:55 PM
Tough to say. What I meant, primarily, is the bogus Middle Eastern turning up in Island people simply because they deviated eastwards relative to the Orkneys cluster. Orcadians are quite drifted so it was not a good proxy population for all English, Irish, Scots, etc.

Thanks, this is exactly what always made me confused with my PF results in the first place. I never knew if they meant it was pull, or if it was because I have Southern Italian/Albanian ancestry, or if it was because I found out I had some Jewish ancestry through DNA testing.

I really like how 23andme seems to be able to sort through all of that for me! However I will concede here that FTDNA made a step in the right direction to have a Jewish Diaspora category. And that they can give me a percentage in that! I was always one of their customers that they "harassed" with the "Where in the Middle East does your ancestry come from?" survey.

seferhabahir
05-04-2014, 04:01 PM
I figured out a way to look at my actual breakdown (which I thought was 99% Jewish Diaspora and 1% European). It is actually 94% Jewish Diaspora, 5% Middle Eastern (Eastern Afroasiatic) and 1% European (European Northlands). Eastern Afroasiatic might be a result of a J2a grandfather and an R2a great-great-grandfather (or just random ancient Middle East remnants).

I had thought the European piece would have been North Mediterranean, based on autosomal Oracle results. I'm taking a guess European Northlands might be a result of my mtDNA being J1c7a, and there's a known branch of J1c7a that shows up in Finns and some Norwegians, although no Scandinavians show up in my mtDNA matches, only Ashkenazi. I believe there are specific mtDNA mutations differentiating Scandinavian J1c7a and Ashkenazi J1c7a. I will have to go look more closely.

Still, it's an improvement over my previously meaningless 100% Middle Eastern designation.

GTC
05-04-2014, 04:44 PM
In answer to a question on FTDNA Forums about the time frame of Population Finder ("Is it 100 years, 1000 years, 2000 years?") Doug McDonald, author of Population Finder among other software tools related to analysis of aDNA, had this to say:


All of these things, whether old Pop finder or new myOrigins, or the other companies and Gedmatch, generate a mishmash of time scales. What happens is that if "recent" works with the list of populations they have, that's what's used. If not, older gets stuffed in, in small amounts. This keeps happening all the way back to Africa.

Alexandre
05-04-2014, 07:44 PM
I am very happy with the results of myOrigins for me. I asked Ricardo to look them up for me, and they came out like this:

88% European (66% North Mediterranean Basin + 22% European Coastal islands)
6% Middle Eastern (6% North African Coastlands)
3% African (3% Niger-Congo Genesis)
3% New World (3% Bering Expansion)

This is much better than the old population finder results, that gave me 70 % western european, 24,3 % Middle Eastern and 5 % African.
I am of portuguese origin, and i have even citizenship of Portugal. I am native from Brazil, so the amerindian and african percentages are probably accurate too.

RyGuy
05-04-2014, 09:41 PM
Going to ask again if someone could look up the results for me (B7682) and my father (318482). Thanks in advance!

RCO
05-04-2014, 10:06 PM
I am very happy with the results of myOrigins for me. I asked Ricardo to look them up for me, and they came out like this:

88% European (66% North Mediterranean Basin + 22% European Coastal islands)
6% Middle Eastern (6% North African Coastlands)
3% African (3% Niger-Congo Genesis)
3% New World (3% Bering Expansion)

This is much better than the old population finder results, that gave me 70 % western european, 24,3 % Middle Eastern and 5 % African.
I am of portuguese origin, and i have even citizenship of Portugal. I am native from Brazil, so the amerindian and african percentages are probably accurate too.

That's typical of a Brazilian of Portuguese origin !

I have
92% European (North Mediterranean Basin 68% + Trans-Ural Peneplain 13% + European Coastal Islands 11%)
3% New World (3% Beringian Expansion)
3% African (2% Niger-Congo Genesis + 1% East African Pastoralists)
1% Middle Eastern (North African Coastlands)

I am mostly of old Portuguese origin in Colonial Brazil and one German Silesian Great-Grandfather to explain the Eastern European Peneplain !

Goldschlager
05-04-2014, 10:07 PM
I definitely see an improvement over the previous 100% ME (my almost 200 years old paper trail provides Eastern European Ashkenazi Jews on all lines). The MyO provides 87 % - Jewish Diaspora, 7%- ME (Anatolian Crossroads) and 6% European (Trans Ural Peneplain). I wonder if the 7 % is somewhat related to my paternal line PF 7580/L 584 which appears to be some pre diaspora Jewish old clade. Another interesting thing I notice: one can sort the matches which I presented with the corresponding categories. I have several which are 100% European (I cannot see the actual subcategory) and I am wondering if this information may not be a clue about where/which line originated from those.

AJL
05-04-2014, 10:19 PM
I wonder if the 7 % is somewhat related to my paternal line PF 7580/L 584 which appears to be some pre diaspora Jewish old clade. Another interesting thing I notice: one can sort the matches which I presented with the corresponding categories. I have several which are 100% European (I cannot see the actual subcategory) and I am wondering if this information may not be a clue about where/which line originated from those.

These both seem quite possible to me.

ADW_1981
05-04-2014, 10:26 PM
Would anyone be so kind to look mine up please? I'm in R1b gateway, DF27, and R1b Atlantic projects.
N48992

Thanks

Stephen1986
05-04-2014, 10:30 PM
Going to ask again if someone could look up the results for me (B7682) and my father (318482). Thanks in advance!

Unless they administer a project that you are part of, they can't see it.

Táltos
05-04-2014, 11:02 PM
https://twitter.com/search?q=%23FamilyTreeDNA&src=hash

Piquerobi
05-05-2014, 12:18 AM
So many years on beta and one would expect a real upgrade, not a fake one. I wish I am wrong but so far I haven't see any reason to celebrate.


After all the hype and years of waiting, people are going to be so disappointed. Maps and cute names and descriptions aren't going to cover the fact that they haven't really changed anything.

RyGuy
05-05-2014, 12:34 AM
Unless they administer a project that you are part of, they can't see it.

My father (kit#: 318482) is in:

Sephardic_Anousim
I2a Y-Haplogroup
Irish Families
Italy DNA Project
Jewish Heritage Proj
Donald Usa (MacDonal Project)
Maryland DNA Project
Middle East
Serio

I (kit#: B7682) am in:

Sephardic_Anousim
Dolan
Italy DNA Project
Jewish Heritage Proj
Donald Usa (MacDonal Project)
Maryland DNA Project
ODwyer
Serio

Anyone who can help I would greatly appreciate it.

RCO
05-05-2014, 01:25 AM
So many years on beta and one would expect a real upgrade, not a fake one. I wish I am wrong but so far I haven't see any reason to celebrate.

What would be your points of divergence, Piquerobi ?

Dr_McNinja
05-05-2014, 02:50 AM
https://twitter.com/search?q=%23FamilyTreeDNA&src=hashFTDNA hired Razib Khan?

Ahaddad
05-05-2014, 03:00 AM
Some admistrator can show my results details? Me and my dad kit 310205 and 298933 thanks

Piquerobi
05-05-2014, 03:32 AM
This poster summed up it well:


After all the hype and years of waiting, people are going to be so disappointed. Maps and cute names and descriptions aren't going to cover the fact that they haven't really changed anything.

I hope I'm wrong, I haven't seen my results yet, but I don't think there will be any substantial change, except for "maps and cute names and descriptions" which "aren't going to cover the fact that they haven't really changed anything".


What would be your points of divergence, Piquerobi ?

Táltos
05-05-2014, 04:32 AM
FTDNA hired Razib Khan?
It sounds like they have him as an ancestry analysis consultant. I'm not Twitter savvy.
The only #hashtags I get are when Jimmy Fallon and Justin Timberlake do their hashtag skits!

RCO
05-05-2014, 09:00 AM
Some admistrator can show my results details? Me and my dad kit 310205 and 298933 thanks

Piquerobi, Ahaddad, I only can observe your new results if you join the "Projeto Brasil", then I can send your results via a private message !

Ahaddad
05-05-2014, 12:53 PM
Piquerobi, Ahaddad, I only can observe your new results if you join the "Projeto Brasil", then I can send your results via a private message !

I will join now, me and my dad account.

Piquerobi
05-05-2014, 02:35 PM
Thanks RCO, but I will wait them. :)


Piquerobi, Ahaddad, I only can observe your new results if you join the "Projeto Brasil", then I can send your results via a private message !

J Man
05-05-2014, 03:36 PM
So if we are admins of a project we can see our results? What if you are only a co-admin?

J Man
05-05-2014, 03:39 PM
I am a member of the Calabria DNA Project, Italy DNA Project, Y-DNA haplogroup J Project, U5b and subclades project and numerous others.

AJL
05-05-2014, 03:49 PM
So if we are admins of a project we can see our results? What if you are only a co-admin?

You should still be able to then.

Ahaddad
05-05-2014, 05:28 PM
New Pop. Finder results
Middle Eastern - 98% (75% Anatolian Crossroads + 21% Eastern Afroasiatic + 1% North African Coastlands)
Central/South Asian - 2% (2% Eurasian Heartland)

Is compatible with Lebanese/Iraqi or Syrians i think

Imadaddin Nasimi
05-05-2014, 07:57 PM
Who can show my new population finder results details? My and my mother's results.

AJL
05-05-2014, 08:24 PM
At a mimimum someone would have to know the project and kit numbers, and then also be an admin of a relevant project.

RCO
05-05-2014, 09:58 PM
Now the North Mediterranean Basin map also covers the Iberian Peninsula to the West (it used to cover only Sicily and Southern Italy), a good correction for the Ibero-descendents !

RCO
05-05-2014, 10:04 PM
Who can show my new population finder results details? My and my mother's results.

You should try to ask the Administrators of your Projects or join a new Project where the Adm. can search your results to see if they are available

Solothurn
05-05-2014, 10:22 PM
Deleted double post!

jeanL
05-05-2014, 10:24 PM
To me the feedback thus far, seems to be that the new algorithm is finding somewhat higher African scores in some participants while lower Native American scores compared to 23andme, and AncestryDNA. Interestingly enough the thing that is contributing to the inflation of the African score isn't higher West African but something called East African Pastoralist. In fact, if one minus the East African Pastoralist scores, it would seem to yield somewhat consistent results for some of the participants, whereas for others 23andme yields a higher West African score, while the new FTDNA thing, yields a higher overall African score, but most of it seems to fall on the African pastoralist score. I should point out, that I am taking mostly about people of Cuban background, and that this phenomenon is from components that are <10%, and it a lot of cases <5%, so maybe, that's what's causing the problems. Also people with no known British ancestry are getting scores in the 10-20% for the European Coastal Islands.

Solothurn
05-05-2014, 10:26 PM
I took a quick peek at mine :)

PF 94% Orc 6% ME.
Now have something roughly like this:
63% European Coastal Islands
28% European Northlands
7% Jewish
1% SW Asian

AncestryDNA give me

Great Britain 34%
Scandinavia 28%
Ireland 20%
Europe West 7%
Italy/Greece 6%
Finland/Northwest Russia 4%
Europe East 1%

23andMe have me as similar but 40+% Unspecified Northern European!

RCO
05-05-2014, 10:50 PM
I have 1% East African Pastoralists and 2% Niger-Congo Genesis (3% African and the same 3% Native American - Bering Expansion). At 23andMe I have 1,4 % SS African, so the proportions are more or less the same. Perhaps the European Coastal Islands can be related to parts of Atlantic Iberia, because I have 11% European Coastal Islands and I only have a distant English ancestor in Rio de Janeiro in 1808 !


To me the feedback thus far, seems to be that the new algorithm is finding somewhat higher African scores in some participants while lower Native American scores compared to 23andme, and AncestryDNA. Interestingly enough the thing that is contributing to the inflation of the African score isn't higher West African but something called East African Pastoralist. In fact, if one minus the East African Pastoralist scores, it would seem to yield somewhat consistent results for some of the participants, whereas for others 23andme yields a higher West African score, while the new FTDNA thing, yields a higher overall African score, but most of it seems to fall on the African pastoralist score. I should point out, that I am taking mostly about people of Cuban background, and that this phenomenon is from components that are <10%, and it a lot of cases <5%, so maybe, that's what's causing the problems. Also people with no known British ancestry are getting scores in the 10-20% for the European Coastal Islands.

jeanL
05-05-2014, 11:02 PM
I have 1% East African Pastoralists and 2% Niger-Congo Genesis (3% African and the same 3% Native American - Bering Expansion). At 23andMe I have 1,4 % SS African, so the proportions are more or less the same. Perhaps the European Coastal Islands can be related to parts of Atlantic Iberia, because I have 11% European Coastal Islands and I only have a distant English ancestor in Rio de Janeiro in 1808 !

That's what I mean, 1.4% SSA becomes 3% African in your case, likewise I've seen people with 2.8%(Most of it West African) SSA in 23andme get 4%(3%East African Pastoralist, 1% Niger-Congo) in here, or people with 1.9% in 23andme get 4% here. East African Pastoralists weren't taken as slaves to the New World, so this component must be representing something else. I mean perhaps one person could indeed have that, but what are the odds that multiple people get that. Also, how does your native compare between 23andme, and here?

faulconer
05-05-2014, 11:10 PM
My results have not changed. Are the results posted under the same pop finder link or something new? I am still 100% Orcadian.

rivergirl
05-05-2014, 11:48 PM
This is how I find the results.
Go into your GAP pages.
Select the "Members Information" page.
There should be a link next to kits with FF results, that says "View".

rivergirl
05-05-2014, 11:58 PM
Now the North Mediterranean Basin map also covers the Iberian Peninsula to the West (it used to cover only Sicily and Southern Italy), a good correction for the Ibero-descendents !

Im pretty sure my European coastal Plains map is larger aswell. It appear to take in more of Germany than before, and extends more eaterley towards CZ, Austria and overlaps with the North Med Basin.

RCO
05-06-2014, 12:26 AM
JeanL, I have 4.2% East Asian & Native American - 3,2% Native American (Standard) at 23andMe and 3% New World Beringian Expansion at My Origins FTDNA, so more or less the same.

Táltos
05-06-2014, 01:24 AM
Now the North Mediterranean Basin map also covers the Iberian Peninsula to the West (it used to cover only Sicily and Southern Italy), a good correction for the Ibero-descendents !
RCO is the North Mediterranean still also over parts of North Africa and the Balkans?

I also just realized in the preliminary results that I have (I think they might change now) the North African Coastlands also was very light green over southern Spain. Has that changed now too?

Someone also posted over at 23andme that Germany is now in two parts. Western=European Coastal Plain. Eastern=Trans Ural. I think they are really started to push a lot into that Trans Ural already.

It's apparent they were not just working out "bugs" with the website, and I really want to know who they used for these samples. I have not seen much posted about what was really said in that Webinar, except that it's release was "imminent". And then it was posted at FTDNA forum something to the effect of well sometime in the next week, which is NOT imminent at all.

Finally and then I'll stop. I really want to know how are they going to compare us with our matches. The FAQ say you will see your top three that you have in common. The reality is most people will not have more than two meaning say Europe and Jewish. Unless what they really mean will they show us that we match our matches because we both have Trans Ural and European Coastal Plains in common? If not this is completely useless. I mean really, who can't tell they probably match someone on the European, or African, or Native American and so on part???

rod
05-06-2014, 02:46 AM
101%, wow, über european ;)
Here's the map.
1824

cyna
05-06-2014, 02:53 AM
They can't even get the math right?

N21163
05-06-2014, 03:51 AM
Rod is just Uber-European! Lol

It's sad to hear that some people don't feel they are getting good results from the MyOrigins update.
I managed to get better results than my original Family Finder.

Family Finder = 95.56% European (Orcadian), 4.44% South Asian (Indian) with 0.73% variance.
1825
MyOrigins results:

96% European
- 58% "European Coastal Islands"
- 29% "European Northlands"
- 7% "North Mediterranean"
- 2% "Trans-Ural Peneplain"

4% Central/South Asian
- 4% "Eurasian Heartland"

1826

Suprised I had no European Coastal Plain in my results...not that I was meant to have any...just curious

vettor
05-06-2014, 03:56 AM
where are all you guys getting this myOrigins data from?

Táltos
05-06-2014, 04:00 AM
Rod is just Uber-European! Lol

It's sad to hear that some people don't feel they are getting good results from the MyOrigins update.
I managed to get better results than my original Family Finder.

Family Finder = 95.56% European (Orcadian), 4.44% South Asian (Indian) with 0.73% variance.
1825
MyOrigins results:

96% European
- 58% "European Coastal Islands"
- 29% "European Northlands"
- 7% "North Mediterranean"
- 2% "Trans-Ural Peneplain"

4% Central/South Asian
- 4% "Eurasian Heartland"

1826

Suprised I had no European Coastal Plain in my results...not that I was meant to have any...just curious
So it does look like they will only tell us that what we have in common with our matches is on the basic level. Most disappointing.

Táltos
05-06-2014, 04:01 AM
where are all you guys getting this myOrigins data from?

If you belong to a DNA project you have to ask the administrator for a screen shot.

Luis
05-06-2014, 04:06 AM
Here's the map.
1824

How can one be 101% of something? That is funny!

GTC
05-06-2014, 04:45 AM
How can one be 101% of something? That is funny!

Presumably it's a rounding error. With a bit more care FTDNA could avoid that problem.

thetick
05-06-2014, 05:53 AM
How can one be 101% of something? That is funny!
Just a very simple rounding error.

85.0% is rounded to 85%
10.5% is rounded to 11%
04.5% is rounded to 05%
---------------------------
100% ....Totals.......101%

N21163
05-06-2014, 06:11 AM
So it does look like they will only tell us that what we have in common with our matches is on the basic level. Most disappointing.

Hi Táltos,

I received this information from a group administrator so I don't know anything beyond my basic map and the population descriptions.
There may be more layers to MyOrigins, but won't know until I have full access

Arran

Táltos
05-06-2014, 12:04 PM
Hi Táltos,

I received this information from a group administrator so I don't know anything beyond my basic map and the population descriptions.
There may be more layers to MyOrigins, but won't know until I have full access

Arran
Hi Arran,
I really hope then that there are more layers to this. That was the other part of this update I was really looking forward to. More ancestry tools, like how 23andme provides several.

GTC
05-06-2014, 12:35 PM
I wonder if they're aware the site is down? I get the impression that the Forums site is not exactly a high visibility thing within FTDNA.

Gerlyons
05-06-2014, 01:27 PM
I wonder if they're aware the site is down? I get the impression that the Forums site is not exactly a high visibility thing within FTDNA.

FTDNA Customer service and IT was notified about 7am (MDT) but I am not sure how long it will be before it will be back up.

Isidro
05-06-2014, 01:52 PM
Now the North Mediterranean Basin map also covers the Iberian Peninsula to the West (it used to cover only Sicily and Southern Italy), a good correction for the Ibero-descendents !

What is the average score for Iberians?...while I wait for mine I am getting this eerie feeling that they solved the Iberian issue by not having Iberian at all, I hope I am wrong but either they put us with what other testing companies call Eastern Mediterranean and FTDNA calls North Med or they will put us a French-German descendants...or a combination of both. I do hope I am wrong on this.

evon
05-06-2014, 02:23 PM
Presumably it's a rounding error. With a bit more care FTDNA could avoid that problem.

This is why its still not released, come on guys, whats up with all the negativity :P

Alexandre
05-06-2014, 03:20 PM
What is the average score for Iberians?...while I wait for mine I am getting this eerie feeling that they solved the Iberian issue by not having Iberian at all, I hope I am wrong but either they put us with what other testing companies call Eastern Mediterranean and FTDNA calls North Med or they will put us a French-German descendants...or a combination of both. I do hope I am wrong on this.

1828

The new North Mediterranean Basin streches to Spain and Portugal, as you can see in this map i got from the FTDNA forums. The old one did not. I am glad about it since i am mostly in this cluster and i am of portuguese origin.




Edit:

Just saw in FTDNA forums that myOrigins will be released TODAY! =]


Edit:

It's out now! \o/

ADW_1981
05-06-2014, 04:56 PM
1828

The new North Mediterranean Basin streches to Spain and Portugal, as you can see in this map i got from the FTDNA forums. The old one did not. I am glad about it since i am mostly in this cluster and i am of portuguese origin.




Edit:

Just saw in FTDNA forums that myOrigins will be released TODAY! =]

The UK/Irish cluster is very strong as well. I wonder why Iberians share more with this cluster rather than the generic European one? Could have recent east to west movements have shifted this older similarity away?

RCO
05-06-2014, 04:59 PM
RCO is the North Mediterranean still also over parts of North Africa and the Balkans?

I also just realized in the preliminary results that I have (I think they might change now) the North African Coastlands also was very light green over southern Spain. Has that changed now too?



That's my map after FTDNA corrected the Iberian heatmap

1829

evon
05-06-2014, 05:10 PM
Its been released :)

My Grandmothers results:

New:
http://i1010.photobucket.com/albums/af228/njal2010/gran-1.png

Old:
http://imageshack.com/a/img36/810/fd2w.jpg

Mine (i used to be 100% Orcadian):
http://i1010.photobucket.com/albums/af228/njal2010/moi-1.png


I think the results might still change, and i suspect my grandmother will regain her Asian %, but it might not be South Asian, but Rather Central Asian in My origins. I was surprised that i got so much Med % compared to the British isles, as i usually score highest there about..

ADW_1981
05-06-2014, 05:13 PM
Is there a way to see your relatives breakdown?

evon
05-06-2014, 05:19 PM
Is there a way to see your relatives breakdown?

Only at a regional level, i can only see my relatives by placement and majority ancestry, most being around 50+ % European, i cant see their minority %..Most of my relatives seem to have opted out of being visible..for now, as i only see a few relatives compared to my overall list...

CelticGerman
05-06-2014, 05:21 PM
here is my result (mainly Northern German ancestry)1830

Caesarea
05-06-2014, 05:27 PM
My Population Finder Ethnic Makeup:

18311832

I am impressed!

evon
05-06-2014, 05:27 PM
here is my result (mainly Northern German ancestry)1830

That would suggest that German ancestry is lumped together with Scandinavia, which seems odd given that have a Continental cluster that would fit better..

ADW_1981
05-06-2014, 05:31 PM
Only at a regional level, i can only see my relatives by placement and majority ancestry, most being around 50+ % European, i cant see their minority %..Most of my relatives seem to have opted out of being visible..for now, as i only see a few relatives compared to my overall list...

Thanks. Yeah, some of them have pins when I click their name, most do not. For some reason my dad has 1% North African, and his profile can see others with Mid-East %. It looks like you can only see minority % if you have some of the same component as well. That doesn't seem to make any logical sense.

Telfermagne
05-06-2014, 05:36 PM
The "My Origins" has been released on my normal profile, it's the same as what I saw via the GAP 2.0 - 100% European Coastal Plain.

I'm curious if there's any other predominantly British Isles folks getting a similar result?

At 23andMe I get 52.2% British & Irish, 11.6% French & German, 3.1 % Scandinavian, 29.1% Nonspecific N. Euro, 1% Iberian, 0.1% Nonspecific S. Euro, 0.5% Ashkenazi, 2.2% Nonspecific Euro, and 0.1% West African. If the threshold for FTDNA is in fact 1% as was mentioned earlier in this thread, then I can see why the Iberian, Ashkenazi and West African is not showing up since they are under 1%. But, I'm not sure why FTDNA is not able to divide my Insular vs. Continental European background.

Sein
05-06-2014, 05:51 PM
It's rather accurate in my case. In fact, this is a massive improvement over my previous results:

Eurasian Heartland=58%
Indian Tectonic Plate=18%
Anatolia/Caucasus=8%
North Circumpolar=7%
Asian Northeast=4%
Eastern Afroasiatic=3%
Bering Expansion=2%

I think the "Indian Tectonic" component is just the "South Indian"-specific component we're used to. In one of Dr_McNinja's ADMIXTURE experiments, I got 18% of a similar component. So the description is fairly accurate, this component is basically a "hybrid", usually an even mix between the first West Eurasian populations to enter South Asia, and ENA South Asian hunter-gatherer populations (or "ASI". This component is usually just 50% West Eurasian, and 50% ASI. But it's a very stabilized, and ancient, hybrid).

For comparison, Dr_McNinja:
Eurasian Heartland=49%
Indian Tectonic=38%
North Circumpolar=6%
Trans-Ural Peneplain=3%
East Asian Coastal Islands=5%

Soulblighter:
Indian Tectonic=58%
Eurasian Heartland=28%
East Asian Coastal Islands=8%
Trans-Ural Peneplain=6%

BMG:
Indian Tectonic=57%
Eurasian Heartland=34%
East Asian Coastal Islands=9%

Compared to Dr_McNinja, Soulblighter, and BMG, I don't have any "East Asian Coastal Islands", and I have the ever present 2% Native American admixture. I've always found this to be very cool. Many Americans of African/European descent (especially of European descent) take DNA tests in hopes of validating family legends about a Native American ancestor. I know that I don't have any Native American ancestry. Yet, almost every test puts me at 2%! Geno 2.0, Dr. McDonald, the various calculators at GEDMatch, and now myORIGINS. I guess it's an excess ANE affinity.

The unfortunate thing, I can't see any results for my matches. All of my matches are either Iranians from Khorasan, or Americans of European descent. But I'm really interested in the results for my eastern Iranian matches. I suspect my results will be closest to them, besides any Pashtuns who've taken this test.

Edit: It's easy to calculate ASI percentages based on these results. The "Indian Tectonic" component is going to be 50% West Eurasian, 50% ASI, as these sorts of clusters are always evenly mixed between West and East Eurasia. The "East Asian Coastal Islands" seems to be excess ASI for South Asians. I don't know why I lack this component. Maybe it only exists east of the Indus? Assuming ASI is 50% of the "Indian Tectonic"+"East Asian Coastal Islands":

Myself=9% ASI

Dr_McNinja=24% ASI

Soulblighter=37% ASI

BMG=37.5%

Seems reasonable.

Stephen1986
05-06-2014, 05:52 PM
I've got 100% Euro, further broken down into 85% European Coastal Islands and 15% European Northlands. My brother has a bit more - he is 98% Euro and 2% Middle Eastern - further broken down into 65% European Coastal Islands, 24% European Coastal Plain, 9% European Northlands and 2% Eastern Afroasiatic.

CelticGerman
05-06-2014, 05:52 PM
That would suggest that German ancestry is lumped together with Scandinavia, which seems odd given that have a Continental cluster that would fit better..

Maybe Northern German is included in Northlands in fact??? See results for my French wife (first) and our four children 18341835183618381839

Humanist
05-06-2014, 05:55 PM
They have renamed "Anatolian Crossroads" to "Anatolia and Caucasus." Apparently, Iraqi Mandaeans have nearly nothing to do with what is today central or southern Iraq. :\ Anyone familiar with their history, language, etc. would recognize this to be an absurdity, of course.

http://i1178.photobucket.com/albums/x372/paulgiva78/passover/mandaean_.jpg

J Man
05-06-2014, 05:55 PM
Mine are now in. I was expecting a higher European Coastal Islands percentage but meh.

North Circumpolar: 30%
North Mediterranean Basin: 29%
European Coastal Islands: 19%
European Coastal Plain: 7%
Trans-Ural Peneplain: 3%
European Northlands: 2%
Eurasian Heartland: 6%
Jewish Diaspora: 5%

evon
05-06-2014, 06:01 PM
Maybe Northern German is included in Northlands in fact??? I will check results of my four children.

https://www.familytreedna.com/learn/ftdna/myorigins-population-clusters/


Northlands & Coastal Plain

As the North European plain aside the North Sea and Baltic region tapers off into the Jutland peninsula Central Europe becomes the North. Where one ends and the other begins is in some ways a matter of debate. Where Denmark begins and Germany ends has been more the outcome of historical coincidence than any geographic or genetic fact. The European Northlands-European Central Plain cluster is a composite which acknowledges that many people bear the hallmarks of straddling these groups, which are genetically very near each other.

People whose ancestors hail from Denmark or the broad plains of Saxony in northern Germany often exhibit affinity with this cluster, due to the recent history of migration south, as well as the deep history of settlement of Scandinavia from Central Europe. This composite is the northern face of the European Coastal Plain, and rightly acknowledges the connection of the German speaking peoples which span different genetic clusters.

As i expected...

AJL
05-06-2014, 06:05 PM
Father:
Jewish Diaspora 83%
Anatolia & Caucasus 10%
North Africa Coastland 2%
European Northlands 5%

Mother:
European Coastal Islands 82%
European Coastal Plain 18%

Seems a fair bit better than Population Finder.

Mehrdad
05-06-2014, 06:05 PM
Here's mine, I'm not that impressed with my maternal side, since they possibly classed the Polynesian/Tongan side as Asian, and Melanesian/Fijian side as Madagascar. And then I get Africa, seriously 3% from Africa?

1837

jeanL
05-06-2014, 06:07 PM
I'm looking forward for them to release their white paper to see which reference populations they used, also what their algorithm for estimating proportions is?

evon
05-06-2014, 06:08 PM
It's rather accurate in my case. In fact, this is a massive improvement over my previous results:

Eurasian Heartland=58%
Indian Tectonic Plate=18%
Anatolia/Caucasus=8%
North Circumpolar=7%
Asian Northeast=4%
Eastern Afroasiatic=3%
Bering Expansion=2%

I think the "Indian Tectonic" component is just the "South Indian"-specific component we're used to. In one of Dr_McNinja's ADMIXTURE experiments, I got 18% of a similar component. So the description is fairly accurate, this component is basically a "hybrid", usually an even mix between the first West Eurasian populations to enter South Asia, and ENA South Asian hunter-gatherer populations (or "ASI". This component is usually just 50% West Eurasian, and 50% ASI. But it's a very stabilized, and ancient, hybrid).

For comparison, Dr_McNinja:
Eurasian Heartland=49%
Indian Tectonic=38%
North Circumpolar=6%
Trans-Ural Peneplain=3%
East Asian Coastal Islands=5%

Soulblighter:
Indian Tectonic=58%
Eurasian Heartland=28%
East Asian Coastal Islands=8%
Trans-Ural Peneplain=6%

Compared to Dr_McNinja and Soulblighter, I don't have any "East Asian Coastal Islands", and I have the ever present 2% Native American admixture. I've always found this to be very cool. Many Americans of African/European descent (especially of European descent) take DNA tests in hopes of validating family legends about a Native American ancestor. I know that I don't have any Native American ancestry. Yet, almost every test puts me at 2%! Geno 2.0, Dr. McDonald, the various calculators at GEDMatch, and now myORIGINS. I guess it's an excess ANE affinity.

The unfortunate thing, I can't see any results for my matches. All of my matches are either Iranians from Khorasan, or Americans of European descent. But I'm really interested in the results for my eastern Iranian matches. I suspect my results will be closest to them, besides any Pashtuns who've taken this test.

I think people of Romani ancestry like my granny will see the South Asian Ancestry divided between Anatolian and Trans-Euro Peneplain, as that is what ive seen so far, i hope they refine the latter, as its too open a category..i agree with you, your results have improved ten fold..so have mine :)

Hok
05-06-2014, 06:08 PM
Cool, my results came in:

100% European:
European Coastal Plain: 69%
European Northlands: 29%
North Circumpolar(Interesting): 3%



Huge improvement, i don't have any European Coastal Islands now(Don't have any known British ancestry) while i was an Orcadian in denial according to the old results. North Circumpolar is a mystery to me tough.

BMG
05-06-2014, 06:08 PM
MY NEW RESULTS

1833

The Indian tectonic plate seems to a south indian component while the eurasian heartland is more similar to baloch in harappa . The east asian is confusing ,i cant say its for real but could be of some tribal ancestry which is a possibility in my case .

Dr_McNinja
05-06-2014, 06:09 PM
They have renamed "Anatolian Crossroads" to "Anatolia and Caucasus." Apparently, Iraqi Mandaeans have nearly nothing to do with what is today central or southern Iraq. :\ Anyone familiar with their history, language, etc. would recognize this to be an absurdity, of course.

http://i1178.photobucket.com/albums/x372/paulgiva78/passover/mandaean_.jpgAll these components from the mideast are very close to Caucasian. Like the Bedouin or Southwest Asian component, it's hard to tell it apart from Caucasian and that's all the way in southwestern Arabia. I mean they could have added Bedouin to this and called it "Anatolia, Caucasus and Middle East" since it's the same as the earlier Middle Eastern component, the main problem was the misleading name.

Simply Southwest Asian was another option. Would work since they have South/Central Asian.

evon
05-06-2014, 06:11 PM
I'm looking forward for them to release their white paper to see which reference populations they used, also what their algorithm for estimating proportions is?

http://i1010.photobucket.com/albums/af228/njal2010/10338519_489737374485078_5258868340042263283_o.jpg
http://i1010.photobucket.com/albums/af228/njal2010/1913471_489746574484158_7949729500743214629_o.jpg
http://i1010.photobucket.com/albums/af228/njal2010/10257467_489746744484141_5378716532381764075_o.jpg

jeanL
05-06-2014, 06:17 PM
http://i1010.photobucket.com/albums/af228/njal2010/10338519_489737374485078_5258868340042263283_o.jpg
http://i1010.photobucket.com/albums/af228/njal2010/1913471_489746574484158_7949729500743214629_o.jpg
http://i1010.photobucket.com/albums/af228/njal2010/10257467_489746744484141_5378716532381764075_o.jpg

Where is that from? Is it from the webinar?

evon
05-06-2014, 06:18 PM
Where is that from? Is it from the webinar?

jupp via FaceBook DNA group person who took screenshots...

Yggdrasil
05-06-2014, 06:19 PM
I really don´t know what to think. I have two kits of full siblings that show more than 40% difference.

Dr_McNinja
05-06-2014, 06:20 PM
Here's the higher res version of mine: http://i.imgur.com/jn51LyU.png (South Asian from Punjab)

Sein
05-06-2014, 06:23 PM
http://i1010.photobucket.com/albums/af228/njal2010/10338519_489737374485078_5258868340042263283_o.jpg
http://i1010.photobucket.com/albums/af228/njal2010/1913471_489746574484158_7949729500743214629_o.jpg
http://i1010.photobucket.com/albums/af228/njal2010/10257467_489746744484141_5378716532381764075_o.jpg

Thanks evon!

I'm one of those 33 Pashtuns.

evon
05-06-2014, 06:23 PM
I really don´t know what to think. I have two kits of full siblings that show more than 40% difference.

Its a crude tool, what kind of difference? you should bring it up at FTDNA's forum, they will likely listen if your results is highly unusual..

Dr_McNinja
05-06-2014, 06:28 PM
Thanks evon!

I'm one of those 33 Pashtuns.If they used Pashtun for Eurasian Heartland and Gujarati for Indian Tectonic, that'd explain a lot. Also why I got drawn outside of the Indian component for the first time since in all my admixture runs I had affinities to HGDP Pathans/Pashtun and Uttar Pradesh but was kind of averse to Gujaratis. The Baloch component/signal is present to similar amounts in both so this kind of reduces its influence.

What were your results like? Did FTDNA post the 23andMe transfers?

CelticGerman
05-06-2014, 06:28 PM
To resume the results of my family (me above all Northern German, my wife French from Burgundy):

Northlands: Me 58%, my wife 0%, our children 18%, 50%, 19%, 33%

Coastal Plain: All 0% besides the third child with 45%

Coastal Islands: Me 13%, my wife 28%, children 29%, 12%, 9%, 9%

North Med: Me 11%, my wife 53%, children 31%, 35%, 5%, 36%

Trans Ural: Me 18%, my wife 18%, children 22%, 0%, 19%, 23%

Anatolia Caucasus: All 0% besides second child with 3%

Eastern Afro Asiatic: All 0% besides third child 4% and my wife with 1%

All children are our children. The differences seem strange to me.

Mehrdad
05-06-2014, 06:29 PM
The African component on my results has me thinking, Siddi somewhere in my ancestry? It'll be pretty cool if it is.

Sein
05-06-2014, 06:32 PM
If they used Pashtun for Eurasian Heartland and Gujarati for Indian Tectonic, that'd explain a lot. Also why I got drawn outside of the Indian component for the first time since in all my admixture runs I had affinities to HGDP Pathans/Pashtun and Uttar Pradesh but was kind of averse to Gujaratis. The Baloch component/signal is present to similar amounts in both so this kind of reduces its influence.

What were your results like? Did FTDNA post the 23andMe transfers?

My results are at page 20, #199. I've also compared your results to my own.

evon
05-06-2014, 06:32 PM
To resume the results of my family (me above all Northern German, my wife French from Burgundy):

Northlands: Me 58%, my wife 0%, our children 18%, 50%, 19%, 33%

Coastal Plain: All 0% besides the third child with 45%

Coastal Islands: Me 13%, my wife 28%, children 29%, 12%, 9%, 9%

North Med: Me 11%, my wife 53%, children 31%, 35%, 5%, 36%

Trans Ural: Me 18%, my wife 18%, children 22%, 0%, 19%, 23%

Anatolia Caucasus: All 0% besides second child with 3%

Eastern Afro Asiatic: All 0% besides my wife with 1%

All children are our children. The differences seem strange to me.

You should start a discussion on the FTDNA forum about this, its clearly an error on their part with regards to your children getting % that the parents dont have...

jeanL
05-06-2014, 06:33 PM
To resume the results of my family (me above all Northern German, my wife French from Burgundy):

Northlands: Me 58%, my wife 0%, our children 18%, 50%, 19%, 33%

Coastal Plain: All 0% besides the third child with 45%

Coastal Islands: Me 13%, my wife 28%, children 29%, 12%, 9%, 9%

North Med: Me 11%, my wife 53%, children 31%, 35%, 5%, 36%

Trans Ural: Me 18%, my wife 18%, children 22%, 0%, 19%, 23%

Anatolia Caucasus: All 0% besides second child with 3%

Eastern Afro Asiatic: All 0% besides my wife with 1%

All children are our children. The differences seem strange to me.

I thought your wife was Half British/Half Greek or something given her results, I would have never guessed she was French. How can she not get any European Coastal Plain while being French, it doesn't make any sense to me!! I also noticed the disparities between some of your children and your results, namely, that one of them gets 101% when adding up everything.

CelticGerman
05-06-2014, 06:36 PM
Thanks evon!

I'm one of those 33 Pashtuns.

Only 17 samples from Germany and France each? Seems ridiculous to me. I would like to know from which regions they are.

vettor
05-06-2014, 06:36 PM
With my newOrigins, I only get

European Coastal at 78% which centres in Western Austria, North Italy, Central-Eastern France, Southern Germany and ALL of Switzerland and other 22% is
Only in Anatolia and Caucasus

Seems to fit my Alpine origins ( i.e later) as stated by DougM many years ago

No Matches

jeanL
05-06-2014, 06:38 PM
Only 17 samples from Germany and France each? Seems ridiculous to me. I would like to know from which regions they are.

Also how many times are they running the ADMIXTURE software, and is this a supervised run, or a side by side run, what's the block size, etc? Now very explicative thus far, hence, why I want to see the white paper.

Dr_McNinja
05-06-2014, 06:39 PM
I thought your wife was Half British/Half Greek or something given her results, I would have never guessed she was French. How can she not get any European Coastal Plain while being French, it doesn't make any sense to me!! I also noticed the disparities between some of your children and your results, namely, that one of them gets 101% when adding up everything.There are areas of France which are genetically more like Spain than Germanic Northwest. It looks like going with the third child's results that it's equating a mix of Coastal Islands and North Med as Coastal Plains. That's the way Admixture works more or less, your admixture can be described as an average of two locations which you're in the middle of.

CelticGerman
05-06-2014, 06:42 PM
Also how many times are they running the ADMIXTURE software, and is this a supervised run, or a side by side run, what's the block size, etc? Now very explicative thus far, hence, why I want to see the white paper.

I would say it's a progress in comparison with Population Finder (100% Orkney and British Islands on the map for me), but it's still something in progress we shouldn't take too seriously.

Humanist
05-06-2014, 06:45 PM
Here's mine, I'm not that impressed with my maternal side, since they possibly classed the Polynesian/Tongan side as Asian, and Melanesian/Fijian side as Madagascar. And then I get Africa, seriously 3% from Africa?

1837

For folks like you (and myself as well) I still trust the results of projects like Harappa when it comes to providing the most meaningful insight into our "Origins."

vettor
05-06-2014, 06:46 PM
There are areas of France which are genetically more like Spain than Germanic Northwest. It looks like going with the third child's results that it's equating a mix of Coastal Islands and North Med as Coastal Plains. That's the way Admixture works more or less, your admixture can be described as an average of two locations which you're in the middle of.

Also the good thing is that they are moving away from the inaccurate modern national boundaries......as an example, how can in the old method that southern France with a latitude of 42degrees be classified as northern European and North Italy with a latitude of 45 degrees ( which is higher ) be classified as southern European..........this is Illogical.

My opinion is not yet finalized on myOrigins........I will await until its bedded down ..........say in a months time.

Mehrdad
05-06-2014, 06:46 PM
I'm with you, I don't have any matches either.

evon
05-06-2014, 06:48 PM
For folks like you (and myself as well) I still trust the results of projects like Harappa when it comes to providing the most meaningful insight into our "Origins."

Harappa totally missed my grandmothers Romani ancestry, so i would not trust any of them alone, best to compare the lot of results and take what correlates across several of them as something worth pursuing in terms of likely ancestry.

One things to keep in mind with My origin is the spill over effect, as many of the clusters are fluid, which could explain the guy with the kid getting Anatolian, which is likely a spill over from Northern Med basin found in his parents...

soulblighter
05-06-2014, 06:50 PM
It's rather accurate in my case. In fact, this is a massive improvement over my previous results:

Eurasian Heartland=58%
Indian Tectonic Plate=18%
Anatolia/Caucasus=8%
North Circumpolar=7%
Asian Northeast=4%
Eastern Afroasiatic=3%
Bering Expansion=2%

I think the "Indian Tectonic" component is just the "South Indian"-specific component we're used to. In one of Dr_McNinja's ADMIXTURE experiments, I got 18% of a similar component. So the description is fairly accurate, this component is basically a "hybrid", usually an even mix between the first West Eurasian populations to enter South Asia, and ENA South Asian hunter-gatherer populations (or "ASI". This component is usually just 50% West Eurasian, and 50% ASI. But it's a very stabilized, and ancient, hybrid).

For comparison, Dr_McNinja:
Eurasian Heartland=49%
Indian Tectonic=38%
North Circumpolar=6%
Trans-Ural Peneplain=3%
East Asian Coastal Islands=5%

Soulblighter:
Indian Tectonic=58%
Eurasian Heartland=28%
East Asian Coastal Islands=8%
Trans-Ural Peneplain=6%

Compared to Dr_McNinja and Soulblighter, I don't have any "East Asian Coastal Islands", and I have the ever present 2% Native American admixture. I've always found this to be very cool. Many Americans of African/European descent (especially of European descent) take DNA tests in hopes of validating family legends about a Native American ancestor. I know that I don't have any Native American ancestry. Yet, almost every test puts me at 2%! Geno 2.0, Dr. McDonald, the various calculators at GEDMatch, and now myORIGINS. I guess it's an excess ANE affinity.

The unfortunate thing, I can't see any results for my matches. All of my matches are either Iranians from Khorasan, or Americans of European descent. But I'm really interested in the results for my eastern Iranian matches. I suspect my results will be closest to them, besides any Pashtuns who've taken this test.



The Bering Expansion might be a 23andMe transfer thing or noise for some people.
In know of another person of my same ethnicity, who has the 1% Bering Expansion as well!


MY NEW RESULTS

1833

The Indian tectonic plate seems to a south indian component while the eurasian heartland is more similar to baloch in harappa . The east asian is confusing ,i cant say its for real but could be of some tribal ancestry which is a possibility in my case .

Looks like all south Indians have coastal east asian.
I wonder if the Coastal East Asian has anything to do with the ASI signature or possibly even ENA.



Here's mine, I'm not that impressed with my maternal side, since they possibly classed the Polynesian/Tongan side as Asian, and Melanesian/Fijian side as Madagascar. And then I get Africa, seriously 3% from Africa?

1837

Very interesting! the 3% could be a Makrani connection from somewhere?

Dr_McNinja
05-06-2014, 06:56 PM
My results are at page 20, #199. I've also compared your results to my own.The only 9% spread between us in Eurasian Heartland makes me curious where other Punjabis would lie. Inbetween us I'd guess since I'm assuming I'd be one of the most eastern-shifted of them. Or they might wind up with more European than us.

mkdexter
05-06-2014, 06:59 PM
The Bering Expansion might be a 23andMe transfer thing or noise for some people.
In know of another person of my same ethnicity, who has the 1% Bering Expansion as well!

Actually I have both FTDNA and 23andMe kits in the system for myself. The transfer was for testing but I can still access it. The transfer is not showing Bering Expansion and of course my FTDNA kit does not either. None of the other transfers in my project have Bering Expansion either. Just thought I'd share that info.

Dr_McNinja
05-06-2014, 07:01 PM
I wonder if they experimented with plotting an actual circle on the map to represent the average of all your components to estimate your geographic origin? The way Dr. McDonald does.

Dr_McNinja
05-06-2014, 07:02 PM
Actually I have both FTDNA and 23andMe kits in the system for myself. The transfer was for testing but I can still access it. The transfer is not showing Bering Expansion and of course my FTDNA kit does not either. None of the other transfers in my project have Bering Expansion either. Just thought I'd share that info.I had a transfer and a FTDNA test and in the old PF they were almost identical.

Sein
05-06-2014, 07:02 PM
The Bering Expansion might be a 23andMe transfer thing or noise for some people.
In know of another person of my same ethnicity, who has the 2% Bering Expansion as well!

That's a very good point. But I think it's a reflection of ANE affinity for South Asians, from South Central Asia all the way to South India. Especially if it occurs repeatedly, using other tools and tests.

I feel pretty lucky though, since I was used in the testing routine, and I am a part of the reference populations. My results might have been less optimal if that wasn't the case.

And the whole Scythian/Sarmatians/Cimmerians part of the description for "Eurasian Heartland" makes sense now, since this component is based primarily on Pashtuns. The component descriptions are very Razib Khan-esque in nature. He tends to describe things using this sort of language, so I wonder if he had a hand in the descriptions? Also, he always emphasizes using the Gujaratis for representation of South Asia-proper, since some samples resemble northwestern Indians/eastern Pakistanis, and some samples resemble "scheduled caste" South Indians. I think he's definitely played a part in designing this.

soulblighter
05-06-2014, 07:03 PM
Actually I have both FTDNA and 23andMe kits in the system for myself. The transfer was for testing but I can still access it. The transfer is not showing Bering Expansion and of course my FTDNA kit does not either. None of the other transfers in my project have Bering Expansion either. Just thought I'd share that info.

Thanks. Are you South Asian also?

warwick
05-06-2014, 07:03 PM
My results were consistent with the 23andme results, but I believe 23andme is better because it paints individual chromosomes and has larger reference populations.

FTDNA results:
European: 66%
Jewish Diaspora: 23% vs 23andme 25.8% Ashkenazi
Middle Eastern: 10%

23andme results:
43.8% Northern European
16.2% Southern European
14.0% non-specific European
Sum of 23andme European: 74% European, 0.2% not specified

So, 23andme: I am 74% European, 25.8% Ashkenazi
FTDNA: 66% European, 23% Jewish, 10% Middle Eastern

Known Ancestry: 75% European, 25% Ashkenazi

Humanist
05-06-2014, 07:11 PM
Harappa totally missed my grandmothers Romani ancestry, so i would not trust any of them alone, best to compare the lot of results and take what correlates across several of them as something worth pursuing in terms of likely ancestry.

I never knew your grandmother had Romani ancestry. Interesting.

Certainly, taking the cumulative record is the most important. However, I do not like the fact that FTDNA is highlighting certain areas as corresponding to these components. According to FTDNA, the genes of the Iraqi Mandaean who I posted results for are overwhelmingly (85%) from a spot on the map that appears to be what is today central Turkey, whilst this individual's attested homeland for what has been a very long time, southern Iraq (and Iran), appears to represent 0% of his "myOrigins."

Many of their traditions can be traced back to Babylonian and Zoroastrian origins, and they once spoke a language related to Babylonian Aramaic, and yet have almost no genetic connection to central/southern Iraq according to FTDNA. I find that highly unlikely.


https://www.youtube.com/watch?v=sNwq8hXbMm0

Dr_McNinja
05-06-2014, 07:11 PM
That's a very good point. But I think it's a reflection of ANE affinity for South Asians, from South Central Asia all the way to South India. Especially if it occurs repeatedly, using other tools and tests.

I feel pretty lucky though, since I was used in the testing routine, and I am a part of the reference populations. My results might have been less optimal if that wasn't the case.

And the whole Scythian/Sarmatians/Cimmerians part of description for "Eurasian Heartland" makes sense now, since this component is based primarily on Pashtuns. The component descriptions are very Razib Khan-esque in nature. He tends to describe things using this sort of language, so I wonder if he had a hand in the descriptions? Also, he always emphasizes using the Gujaratis for representation of South Asia-proper, since some samples resemble northwestern Indians/eastern Pakistanis, and some samples resemble "scheduled caste" South Indians. I think he's definitely played a part in designing this.I think they kind of got lucky with how well Eurasian Heartland turned out. Baloch seems like a more natural component to use but it causes some chaos in the immediate vicinity as Pashtuns (distinct from Pakistan-area Pathans) have less Baloch and northern Indians also have less Baloch despite being surrounded by people with lots of Baloch. The presence of Baloch in the old PF kind of pushed the Pashtun out of the picture entirely into the Middle East component. It also was leaking out some European since I was registering Orcadian in the old PF. But with using Pashtun as the Northwest Indian proxy, everyone will have an affinity to them just as well as they did to the Baloch and it draws in the Indo-European admixture pretty well, which Baloch did not do. So much so that now Europeans are registering some of it as they registered Gedrosian in admixture calculators. And that link (Eurasian Heartland-Gedrosian-Orcadian) is probably the Sarmatian/Scythian link.

The other thing they could have done was just use a lot more components in the old system to make it work better, but I guess they wanted a more streamlined and easier to understand picture.

vettor
05-06-2014, 07:11 PM
My results were consistent with the 23andme results, but I believe 23andme is better because it paints individual chromosomes and has larger reference populations.

European: 66%
Jewish Diaspora: 23% vs 23andme 25.8% Ashkenazi

Middle Eastern: 10%

23andme European:
23andme 43.8% Northern European
16.2% Southern European
14.0% non-specific European
Sum of 23andme European: 74% European, 0.2% not specified

So, 23andme: I am 74% European, 25.8% Ashkenazi
FTDNA: 66% European, 23% Jewish, 10% Middle Eastern

Known Ancestry: 75% European, 25% Ashkenazi


take note that its larger reference data is "borrowed" from other companies, like 1000genomes etc.
What say does 23andme have for these borrowed reference data ?

warwick
05-06-2014, 07:14 PM
take note that its larger reference data is "borrowed" from other companies, like 1000genomes etc.
What say does 23andme have for these borrowed reference data ?

I still think GEDMATCH tools are the best.

Dr_McNinja
05-06-2014, 07:14 PM
My results were consistent with the 23andme results, but I believe 23andme is better because it paints individual chromosomes and has larger reference populations.

FTDNA results:
European: 66%
Jewish Diaspora: 23% vs 23andme 25.8% Ashkenazi
Middle Eastern: 10%

23andme results:
43.8% Northern European
16.2% Southern European
14.0% non-specific European
Sum of 23andme European: 74% European, 0.2% not specified

So, 23andme: I am 74% European, 25.8% Ashkenazi
FTDNA: 66% European, 23% Jewish, 10% Middle Eastern

Known Ancestry: 75% European, 25% AshkenaziFTDNA is going for older timeframes so that's actually a decent correlation. Middle Eastern is describing a Caucasus-like component and when you go back far enough there is plenty of Caucasian in Europeans. 23andMe cheats and labels everything European in Europeans, and will label the exact same DNA as Middle Eastern in a Middle Eastern person.

warwick
05-06-2014, 07:17 PM
FTDNA is going for older timeframes so that's actually a decent correlation. Middle Eastern is describing a Caucasus-like component and when you go back far enough there is plenty of Caucasian in Europeans. 23andMe cheats and labels everything European in Europeans, and will label the exact same DNA as Middle Eastern in a Middle Eastern person.

Sure but:

1. 23andme assigns results to European nationalities, i.e. Italian, French, etc, which can be more useful
2. 23andme paints the chromosomes
3. 23andme has larger reference populations and more populations

Sein
05-06-2014, 07:19 PM
I think they kind of got lucky with how well Eurasian Heartland turned out. Baloch seems like a more natural component to use but it causes some chaos in the immediate vicinity as Pashtuns (distinct from Pakistan-area Pathans) have less Baloch and northern Indians also have less Baloch despite being surrounded by people with lots of Baloch. The presence of Baloch in the old PF kind of pushed the Pashtun out of the picture entirely into the Middle East component. It also was leaking out some European since I was registering Orcadian in the old PF. But with using Pashtun as the Northwest Indian proxy, everyone will have an affinity to them just as well as they did to the Baloch and it draws in the Indo-European admixture pretty well, which Baloch did not do. So much so that now Europeans are registering some of it as they registered Gedrosian in admixture calculators. And that link (Eurasian Heartland-Gedrosian-Orcadian) is probably the Sarmatian/Scythian link.

The other thing they could have done was just use a lot more components in the old system to make it work better, but I guess they wanted a more streamlined and easier to understand picture.

I think they took some recommendations from geneticists with the Baloch/Brahui. The Baloch/Brahui are much more drifted than Pashtuns, so I think they thought using Pashtuns was a better option. In pragmatic terms, it's a big improvement over me being almost completely "Middle Eastern"! I wonder how Baloch/Brahui individuals will look, using this system?

Just out of anthropological curiosity, who are Pathans? Do you mean North Indians who have distant Pashtun roots, and live in Pakistan? Or acculturated groups in Punjab, like the Niazi, who don't speak Pashto, and are predominantly Punjabi in ancestry due to intermarriage, but have distant Pashtun roots?

mnd
05-06-2014, 07:22 PM
This seems to be an improvement over Population Finder. I transferred my maternal grandmother’s results from 23andMe and was given this breakdown:

86.28% Middle Eastern (Druze)
13.72% European (Finnish, Orcadian, Russian)

Her myOrigins results pick up on her minor East African ancestry, don't suggest that she is of mostly Druze ancestry, and give a more comprehensible breakdown:

77% Middle Eastern:
- 57% Anatolia and Caucasus
- 17% Eastern Afroasiatic
- 3% North African Coastlands

21% European:
- 21% North Mediterranean Basin

3% African:
- 3% East African Pastoralists

warwick
05-06-2014, 07:25 PM
Sure but:

1. 23andme assigns results to European nationalities, i.e. Italian, French, etc, which can be more useful
2. 23andme paints the chromosomes
3. 23andme has larger reference populations and more populations

It is fair to say that each tool whether GEDMATCH, 23andme, or FTDNA gives different results based on certain reasonable assumptions and the different size of the samples used. Still, I think GEDMATCH is the best of the available options, given the variety of choices.

i.e. within limits, each set of results should be viewed provisionally in light of known ancestry and the available sequences and algorithms used for the analyses.