U.S. CitiesCity-Data Forum Index
Covid-19 Information Page
Go Back   City-Data Forum > General Forums > Genealogy
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
Reply Start New Thread
 
Old 08-03-2017, 11:16 PM
 
Location: CA--> NEK VT--> Pitt Co, NC
379 posts, read 314,245 times
Reputation: 418

Advertisements

Quote:
Originally Posted by Alandros View Post
So autosomal DNA doesn't have haplogroups, they summarize the results there by assigned ethnic percentages.

Haplogroups are simply the way they summarize what Y DNA, paternal only DNA, matches in relation to the rest of the world. Since it only represents one line they don't do percentages, they do designations. Also because of the limitations you point out regarding autosomal DNA since it is a mixed bag of all your DNA it can't tell you too much detail further back except rough ethnic groups you likely match. Y DNA can match any other male around the world at some level, whether 200 years ago or 10,000 years ago.

Because your dad was assigned to Haplogroup L-21 that only applies to his paternal Y DNA, it does not apply to the autosomal DNA results. That means that paternally you and he descend from a European male, L21 being the "Atlantic Celtic" haplgroup, the most common in Ireland, Scotland, and England... but found at lower levels outside of it such as France or Germany.

So again L21 only applies to the Y DNA, no other part of the DNA and it is not a summary for all your DNA results.
Per ISOGG:

"Y-chromosome DNA (Y-DNA) haplogroups are determined by single-nucleotide polymorphism (SNP) tests. SNPs are locations on the DNA where one nucleotide has "mutated" or "switched" to a different nucleotide.

Because a haplogroup consists of similar haplotypes, it is possible to predict a haplogroup from the haplotype. A SNP test is required to confirm the haplogroup prediction"



Are you saying 23 & Me does not do SNP tests? If so, how do they end up giving a result with a haplogroup? From what I understand, some vendors provide SNP tests with autosomal DNA test; some don't. I was under the impression that 23 & Me was in the former group.

Is that not true?
Reply With Quote Quick reply to this message

 
Old 08-03-2017, 11:30 PM
 
877 posts, read 893,996 times
Reputation: 1227
Quote:
Originally Posted by naadarien View Post
Per ISOGG:

"Y-chromosome DNA (Y-DNA) haplogroups are determined by single-nucleotide polymorphism (SNP) tests. SNPs are locations on the DNA where one nucleotide has "mutated" or "switched" to a different nucleotide.

Because a haplogroup consists of similar haplotypes, it is possible to predict a haplogroup from the haplotype. A SNP test is required to confirm the haplogroup prediction"



Are you saying 23 & Me does not do SNP tests? If so, how do they end up giving a result with a haplogroup? From what I understand, some vendors provide SNP tests with autosomal DNA test; some don't. I was under the impression that 23 & Me was in the former group.

Is that not true?
So one thing to note is autosomal SNPs are different than Y DNA SNPs, basically they both have markers roughly called SNPs... Again it's important to think of it as different types of DNA tested, each type of DNA is separate and contains it's own type of information.

As far as determining a haplogroup they look at the Y DNA and compare it to the known SNPs (or STRs to some extent can be used as well) that are known to be common to a certain haplogroup. Basically they look at the Y DNA signature and tell you which haplogroup that signature fits.

23andme only provides limited Y DNA haplogroup information, they test a certain amount of SNPs and no more. At a place like Family Tree DNA (FTDNA) you can have more extensive tests done, testing specific SNPS or what is now becoming more common (though still relatively expensive) a comprehensive Y DNA test that tests a huge portion of SNPs. So if you test at FTDNA you'll get a haplogroup as well, but can test deeper.

In this case it's important to know that haplogroups are a tree, so there are levels below L21 with specific sets of SNPS that represent the sub-branches. New SNP markers are still being discovered frequently and the branches expand all the time.
Reply With Quote Quick reply to this message
 
Old 08-03-2017, 11:53 PM
 
Location: Illinois
3,168 posts, read 4,470,245 times
Reputation: 5596
Quote:
Originally Posted by naadarien View Post
Cool. So how is that haplogroup determined?
It combs through your direct maternal line (if you are female) and direct maternal and paternal line (if you are male).

From what I know, 23andMe is the best thus far when it comes to haplogrouping. I also got haplogroup results from James Lick's calculator and I believe We Gene.


There is something interesting between the two results and 23andMe but again, what is your real question?
Reply With Quote Quick reply to this message
 
Old 08-04-2017, 05:09 AM
 
Location: CA--> NEK VT--> Pitt Co, NC
379 posts, read 314,245 times
Reputation: 418
Quote:
Originally Posted by Alandros View Post
23andme only provides limited Y DNA haplogroup information, they test a certain amount of SNPs and no more. At a place like Family Tree DNA (FTDNA) you can have more extensive tests done, testing specific SNPS or what is now becoming more common (though still relatively expensive) a comprehensive Y DNA test that tests a huge portion of SNPs. So if you test at FTDNA you'll get a haplogroup as well, but can test deeper.
Now that might explain the "problem." If they only analyze limited SNPs, then answer could be limited. I assume the answer is really a lot more technical (which is what I would prefer), but this will do for now.

I already asked my father to take the FTDNA y-DNA test. I am interested in seeing how the results differ...or not. He also finally loaded his test to GEDMatch so I can look at some more of the technical data from there.

Thanks.
Reply With Quote Quick reply to this message
 
Old 08-04-2017, 05:23 AM
 
Location: CA--> NEK VT--> Pitt Co, NC
379 posts, read 314,245 times
Reputation: 418
Quote:
Originally Posted by CMichele View Post
It combs through your direct maternal line (if you are female) and direct maternal and paternal line (if you are male).

From what I know, 23andMe is the best thus far when it comes to haplogrouping. I also got haplogroup results from James Lick's calculator and I believe We Gene.


There is something interesting between the two results and 23andMe but again, what is your real question?
I am not sure why we are not communicating, but "how a haplogroup is determined?" is essentially the question I am asking. Knowing how the process actually, technically works, helps explain and validate his results. The unfortunate reality is there isn't a lot of data out there to explain what I can only assume is a highly technical process. It is going to require that I dig into educational materials, and NIH studies, and human genome project documentation to find what that methodology is.

I will confess that I had hoped some genetic scientist had figured out how to distill that information into some description that could be consumed by the masses, but efforts into finding out this methodology has run into pretty much the same brick wall I see here. The vast majority of people don't really know how haplogroups are determined because that isn't a commonly shared detail, and merely saying that haplogroups are determined by deriving haplotypes from a few SNPs has clearly been plenty of info for the lion share of people.

I was always that kid who asked why, why, why until I bugged people to death, so that answer doesn't cut it. I am also an engineer and at the root of it all, I like knowing HOW things work, and it never really has mattered how technical the answer is. I like knowing it. There is likely an explanation to how ALL haplogroups are generally developed for y-DNA, and when I find it, I'll be happy to post it. Someone else must have asked this question once or twice. I'm not special in the question asking department.

Thanks for your help.
Reply With Quote Quick reply to this message
 
Old 08-04-2017, 06:42 AM
 
5,622 posts, read 7,668,504 times
Reputation: 5578
Quote:
Originally Posted by naadarien View Post
You're missing the point.
I want to know more about my dad's main haplogroup. The group that involves the main aspects of his heritage, not the haplogroup of the cosmic dust that has passed down from hundreds of 4 and 5G-grandparents.

That doesn't make sense?

So maybe the better question is... how is the haplogroup assigned? And how could it be that one group would be dominant even if that group is assigned to the smaller/est share of one's chromosomes? There has to be some methodology to how that works. What is that?
The test they can do is very specific. It is one specific line of ancestry - it is father's father's father's father's father's father - how ever far back. It does not tell you about any other line that has mother anywhere in it. And the mtDNA is mother's mother's mother's mother's mother's mother's mother - however far back. The purpose of these tests is not to tell you about the 'main body' of ancestry, but those specific lines only - the reason why is that those particular lines happen to be particularly easy to test - the others are not.
Reply With Quote Quick reply to this message
 
Old 08-04-2017, 07:03 AM
 
Location: CA--> NEK VT--> Pitt Co, NC
379 posts, read 314,245 times
Reputation: 418
Quote:
Originally Posted by otowi View Post
The test they can do is very specific. It is one specific line of ancestry - it is father's father's father's father's father's father - how ever far back. It does not tell you about any other line that has mother anywhere in it. And the mtDNA is mother's mother's mother's mother's mother's mother's mother - however far back. The purpose of these tests is not to tell you about the 'main body' of ancestry, but those specific lines only - the reason why is that those particular lines happen to be particularly easy to test - the others are not.
Yes, and I suspect that that is why some tests seem wildly disparate.

It isn't that it isn't accurate that my dad had a euro ancestor at the end of x chain of fathers, it is that it was easier and less costly to give that answer. Partly because there aren't limitless resources to chase down everything and partly because there isn't reference data for everything. Likely the algorithm is looking for a weighted match, and the moment that is found, the haplogroup is determined. The problem is the "weight" is based on reference data, and the largest contributors to DNA databases are people of european descent (so more reference data). Or something along those lines...I have more research to do.

We had numerous segments in our results that were unable to be identified and one explanation for that is that there haven't been enough DNA tests done to properly identify the segment. There are some really interesting papers written on that subject (just the first of many rabbit holes I have been down since I got my DNA results) BTW.

Hopefully, if I take the test in 20 years, there will be more complete data. I encourage everyone to get tested. :-)
Reply With Quote Quick reply to this message
 
Old 08-04-2017, 07:29 AM
 
Location: Ozark Mountains
634 posts, read 540,190 times
Reputation: 787
There is a country in Central Africa with the same Haplogroup, so not necessarily your great grand parents came from England.
Reply With Quote Quick reply to this message
 
Old 08-04-2017, 07:41 AM
AFP
 
7,241 posts, read 4,713,343 times
Reputation: 6211
Quote:
Originally Posted by naadarien View Post
Say what now?

Seriously. I think 23 & Me has lost their minds.

My dad is 81% SSA, 17.1% Euro and the rest native american.

His father's dead, but his mother's DNA test came up as 83% SSA so his dad has to be in that ballpark too.

I have tried to get them to explain to me how it is that my dad was assigned the R-L21 haplogroup when both of his parents are american blacks and greater than 80% SSA, but nothing they are telling me is making sense. His maternal haplogroup BTW was L3f1b which matches my grandmother's results perfectly.

Since there is no question of my father's parentage, LOL, could someone take a stab at explaining why it worked out this way?

Don't get me wrong. I do think that haplogroup is accurate for that 17% sliver in my dad's result, but I can't understand how that one piece is the dominant paternal haplogroup.

Any ideas?
One third of black Americans have a Y-DNA that has origins in Europe. Your result is quite common the SSA percentage is not at all related to your Y-DNA result. It is possible for someone to have an estimate of 100% SSA and a Y-DNA that has origins in Europe and for someone to have an estimate of 100% European and have an mtDNA or Y-DNA with origins in Sub-Saharan Africa I've seen it. Do some more reading on Y-DNA it's the direct paternal line nothing else.
Reply With Quote Quick reply to this message
 
Old 08-04-2017, 07:46 AM
 
Location: CA--> NEK VT--> Pitt Co, NC
379 posts, read 314,245 times
Reputation: 418
Quote:
Originally Posted by ozarknation View Post
There is a country in Central Africa with the same Haplogroup, so not necessarily your great grand parents came from England.
The narrative that goes with that map: "R1b is the most common haplogroup in Western Europe, reaching over 80% of the population in Ireland, the Scottish Highlands, western Wales, the Atlantic fringe of France, the Basque country and Catalonia. It is also common in Anatolia and around the Caucasus, in parts of Russia and in Central and South Asia. Besides the Atlantic and North Sea coast of Europe, hotspots include the Po valley in north-central Italy (over 70%), Armenia (35%), the Bashkirs of the Urals region of Russia (50%), Turkmenistan (over 35%), the Hazara people of Afghanistan (35%), the Uyghurs of North-West China (20%) and the Newars of Nepal (11%). R1b-V88, a subclade specific to sub-Saharan Africa, is found in 60 to 95% of men in northern Cameroon."

That last sentence is most salient. My father's subclade isn't V88, though I am willing to withhold judgment on that until his FTDNA test comes back.

Also his ancestry showed no Cameroon, and only 1% of his 81% SSA ancestry is central african. The rest is western, and even that is almost exclusively Nigerian and Senegalese.

African topography being what it is, people on the far western edge of the continent tended to go north and east, around the saharan after it became a desert. My father's senegalese DNA reflects that exact situation. He has small bits of Malian and Egyptian DNA because of that migration pattern. So that is a migratory pattern that goes away from the Central African countries.

Again...possible, anything is possible at this point, but until I have evidence that he is rare (beyond being my lovely father), I'm going to bet that he is just average.

ETA: Though, for speculation purposes, I think we are looking at this map backwards. It may well be that all of these different european and asian groups that are associated with this haplogroup actually originate in Cameroon, and then migrated out from there. It would be interesting to study that V88 subclade and see what characteristics of it impacted the characteristics of the other subclades. I could go on endlessly about whether or not we have accurately evaluated the data we have in genetics (but I do that with all kinds of scientific theories). So much is missing, but still it can make for interesting debates. :-)

Last edited by naadarien; 08-04-2017 at 08:06 AM..
Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply
Please update this thread with any new information or opinions. This open thread is still read by thousands of people, so we encourage all additional points of view.

Quick Reply
Message:

Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Genealogy
Follow City-Data.com founder on our Forum or

All times are GMT -6.

© 2005-2020, Advameg, Inc. · Please obey Forum Rules · Terms of Use and Privacy Policy · Bug Bounty

City-Data.com - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35 - Top