Most Recent Common Ancestor 12 Marker Test

[Learning_Genetics]

1. If you take a 12 marker DNA test and get an exact 12/12 match, first of all, how likely is it that the match is authentic?

2. And secondly how long ago can we assume when the most recent common ancestor was in such a case?

3. If you take a 12 marker DNA test and get 11/12 match, how likely is it that the match is authentic.

4. How long ago can we assume the most recent common ancestor was in such a case?

5. Is there a difference between when such tests were conducted? For example if you took the FTDNA 12 marker test in 1999 or 2000, for example, would getting 12/12 match and 11/12 increase reliability or have different implications in terms of most recent common ancestor than if you took it in 2019?

[rms2]

If your exact 12-marker match has your surname, you should pay attention and try to get that person to upgrade.

Otherwise, anything goes with 12-marker matches. Most of them aren't worth paying attention to, but you never know until the person upgrades to at least 67 markers.

[rms2]

I'm not a big fan of Rebecca Canada, but this graphic is generally pretty reliable.

STR matches and relatedness_genetic distance_dna.jpg

[Learning_Genetics]

If your exact 12-marker match has your surname, you should pay attention and try to get that person to upgrade.

Otherwise, anything goes with 12-marker matches. Most of them aren't worth paying attention to, but you never know until the person upgrades to at least 67 markers.

Thank you for your insight.

On the chart you posted it mentions 1 to 15 generations. However this is not really a long time.

On some forums I have seen it mentioned that 12 marker tests can produce matches where last common ancestor was 5,000 to 3,000 years ago.

If it is a 11/12 or 12/12 match on 12 marker test how recent would the last common ancestor be?

If it is even less close match how far would it be?

[msmarjoribanks]

Thank you for your insight.

On the chart you posted it mentions 1 to 15 generations. However this is not really a long time.

On some forums I have seen it mentioned that 12 marker tests can produce matches where last common ancestor was 5,000 to 3,000 years ago.

If it is a 11/12 or 12/12 match on 12 marker test how recent would the last common ancestor be?

If it is even less close match how far would it be?

It probably makes a difference what your haplogroup is. My dad is a perfect match for one of the Western Atlantic Modal Haplotypes at 12 markers, and so has 5800 exact 12-marker matches (nearly 10,000 matches overall).

However, his actual haplogroup is pretty rare (a subclade of R1b-DF63), and so he has only 1 111-marker match (also a BigY match), only 2 other 67 marker matches (distances of 6 and 7) and only 3 37 marker matches (the only perfect match at 37 is the BigY match). Even at 25, he has only 26 matches, a huge drop from 12. He also has no matches in his surname project, and his closest match in the the L21 project (who is also at YFull) has an estimated MRCA date of 3500 YBP. I need to convince the BigY match to upload at YFull.

So perfect 12 matches are often extremely distant. Like rms, I'd pay attention if they share your surname, especially if it's not so common.

[Saetro]

It probably makes a difference what your haplogroup is. My dad is a perfect match for one of the Western Atlantic Modal Haplotypes at 12 markers, and so has 5800 exact 12-marker matches (nearly 10,000 matches overall).

However, his actual haplogroup is pretty rare (a subclade of R1b-DF63), and so he has only 1 111-marker match (also a BigY match), only 2 other 67 marker matches (distances of 6 and 7) and only 3 37 marker matches (the only perfect match at 37 is the BigY match). Even at 25, he has only 26 matches, a huge drop from 12. He also has no matches in his surname project, and his closest match in the the L21 project (who is also at YFull) has an estimated MRCA date of 3500 YBP. I need to convince the BigY match to upload at YFull.

So perfect 12 matches are often extremely distant. Like rms, I'd pay attention if they share your surname, especially if it's not so common.

My Haplogroup T is fairly rare - even moreso for the subclade.
If I find someone with my surname who matches closely at 12 markers, they will likely be a close match in time.
As surnames were only taken on around 700-800 years ago, any such match will be within that time scale, and probably less than half that.

[GoldenHind]

The answer depends on how rare or common your 12 marker results are. If they are common, such as with the WAMH, 12 marker matches are likely to include both people who are closely related as well as many who have no connection within a millennium or two. If you have a rare 12 marker signature, 12/12 marker matches and even 11/12 are more likely to be significant.

[karwiso]

See here: https://www.familytreedna.com/faq-markers.aspx

PROBABILITY FOR MOST RECENT COMMON ANCESTOR (MRCA)
Number of matching markers 95% probability that the MRCA was no longer than this number of generations
10 of 10 72
11 of 12 47
12 of 12 29

As you can see, Y-12 is a pretty useless test unless you have some serious suspicions that this relative wouldn't fit in your family tree.
And https://www.familytreedna.com/learn/...y-dna-matches/

If you compare the 12 marker result to someone else who does not have the same surname, but the scores match, you are most likely NOT recently related. When we use the term recently related, we are talking about a time frame within the last 1000 years or 40 generations, a time depth that accommodates the earliest known use of surnames.

Consider Y-67.

65 of 67 14
66 of 67 9
67 of 67 6

Here you can see much closer matches, or possible relatives, within the timeline of the genealogical/archival records. And here you can find some interesting reading about Y-111: https://www.familytreedna.com/learn/...s-interpreted/

[Learning_Genetics]

Thank you everyone for your help.

It probably makes a difference what your haplogroup is. My dad is a perfect match for one of the Western Atlantic Modal Haplotypes at 12 markers, and so has 5800 exact 12-marker matches (nearly 10,000 matches overall).

However, his actual haplogroup is pretty rare (a subclade of R1b-DF63), and so he has only 1 111-marker match (also a BigY match), only 2 other 67 marker matches (distances of 6 and 7) and only 3 37 marker matches (the only perfect match at 37 is the BigY match). Even at 25, he has only 26 matches, a huge drop from 12. He also has no matches in his surname project, and his closest match in the the L21 project (who is also at YFull) has an estimated MRCA date of 3500 YBP. I need to convince the BigY match to upload at YFull.

So perfect 12 matches are often extremely distant. Like rms, I'd pay attention if they share your surname, especially if it's not so common.

Thank you for your insight.

The answer depends on how rare or common your 12 marker results are. If they are common, such as with the WAMH, 12 marker matches are likely to include both people who are closely related as well as many who have no connection within a millennium or two. If you have a rare 12 marker signature, 12/12 marker matches and even 11/12 are more likely to be significant.

Thank you for explaining.

See here: https://www.familytreedna.com/faq-markers.aspx


As you can see, Y-12 is a pretty useless test unless you have some serious suspicions that this relative wouldn't fit in your family tree.
And https://www.familytreedna.com/learn/...y-dna-matches/


Consider Y-67.


Here you can see much closer matches, or possible relatives, within the timeline of the genealogical/archival records. And here you can find some interesting reading about Y-111: https://www.familytreedna.com/learn/...s-interpreted/

Thank you for the information.

[RobertCasey]

The FTDNA genetic distance charts are extremely unreliable. FTDNA assumes genetic distance are normal distributions. For around a dozen haplogroups that I
have analyzed, the measurement for Normal Distribution fails the statistical measurement - Kurtosis (which always exceeds the normalized value of two). Our
genetic distance curves have fat tails on the left and are skewed to the right in almost every haplogroup analyzed (over twenty).

Here is a presentation that goes over this analysis (you need some understanding of statistics):

https://www.youtube.com/watch?v=pmv7RyrVTa4