PDA

View Full Version : Z36 STR Anomaly



emmental
07-04-2013, 04:10 PM
While waiting for my Full Y Gemome results I’m keeping my mind occupied by playing around with Z36 STRs. I took all the Z36 surnames and searched the projects to get as accurate a haplotype for each surname as possible. I then combined the close clusters (i.e. Musselman/Binkley/Singer; Erskine/White/Stuart/Tyson; and Brandeberry/Barthelmus) to count as only one haplotype for each. This gave me 56 Haplotypes for Y-12; 55 for Y-25; 54 for Y-37; 48 for Y-67; and 11 for Y-111. While tabulating a Z36 Modal haplotype I noticed this anomaly:

R1b frequency for DYS 437 http://freepages.genealogy.rootsweb.ancestry.com/~geneticgenealogy/yfreq.htm
14-12%; 15-85%; 16-3%
Z36 frequency for DYS437
14-42%; 15-53%; 16-5%

R1b frequency for H4
http://freepages.genealogy.rootsweb.ancestry.com/~geneticgenealogy/yfreq.htm
10-20%; 11-71%; 12-9%
Z36 frequency for H4
10-48%; 11-50%; 12-2%

The modal for both holds, but just barely. But here’s the weird thing. You would expect the DYS437=15/H4=11 group to be the largest, since that is the modal. But it isn’t even close. This is how it breaks down:

DYS437=15 & H4=10 21 of 54 or 39%
DYS437=14 & H4=11 19 of 54 or 35%
DYS437=15 & H4=11 8 of 54 or 15%
DYS437=16 & H4=10 3 of 54 or 6%
DYS437=14 & H4=10 2 of 54 or 4%
DYS437=14 & H4=12 1 0f 54 or 2%

The off-modal DYS437 holds the H4 modal and the off-modal H4 holds the DYS437 modal.

Wow! After typing that last sentence I thought to myself “What’s so strange about that?”

Maybe this is just evidence of two early mutations under Z36; DYS437=15>14 and H4=11>10.

Thoughts?

(BTW If you would like to see my spreadsheet, send a message with your email)

Rathna
07-04-2013, 04:37 PM
Maybe this is just evidence of two early mutations under Z36; DYS437=15>14 and H4=11>10.

Thoughts?

It is the same case of DYS426=11 in R-M269* and =13 in R-L51*. Probably the most diffused value was 12, but the mutation M269 or L51 may have happened in a haplotype which had had that mutation, even though not the most diffused. From there that value was the "modal". You should add that the mutations happened around the modal, there is a convergence to the modal as time passes and only sometime a mutation goes for the tangent. But also about this every marker has its rules and a different probability to mutate upward or downward etc etc.

emmental
07-04-2013, 05:33 PM
I guess the strange thing is the lack of what should be the modal; i.e. 437=15/H4=11.
If you would group the outlier values (437=16 and H4=12) back a single step, as opposed to a double step, it would look like this:

DYS437=15/16 & H4=10 24 of 54 or 44%
DYS437=14 & H4=11/12 20 of 54 or 37%
DYS437=15 & H4=11 8 of 54 or 15%
DYS437=14 & H4=10 2 of 54 or 4%

That's 85% who would be off-modal for the duel modal! The double off-modal 14/10 is what you'd expect (4%) but the rest seems to point to something that needs explaining. Eighty percent of Z36 is either DYS437=14 or H4=10.

Rathna
07-04-2013, 05:56 PM
I guess the strange thing is the lack of what should be the modal; i.e. 437=15/H4=11.
If you would group the outlier values (437=16 and H4=12) back a single step, as opposed to a double step, it would look like this:

DYS437=15/16 & H4=10 24 of 54 or 44%
DYS437=14 & H4=11/12 20 of 54 or 37%
DYS437=15 & H4=11 8 of 54 or 15%
DYS437=14 & H4=10 2 of 54 or 4%

That's 85% who would be off-modal for the duel modal! The double off-modal 14/10 is what you'd expect (4%) but the rest seems to point to something that needs explaining. Eighty percent of Z36 is either DYS437=14 or H4=10.

Probably you don't calculate that these markers have different mutation rate and that mutations happened forwards and backwards and that some markers mutate around a few values, like DYS391 which rotates above all around the values 10 and 11 and only rarely happens the mutation for the tangent. This brings you to neglect hidden mutations.

emmental
07-04-2013, 06:15 PM
Probably you don't calculate that these markers have different mutation rate and that mutations happened forwards and backwards and that some markers mutate around a few values, like DYS391 which rotates above all around the values 10 and 11 and only rarely happens the mutation for the tangent. This brings you to neglect hidden mutations.

The DYS 391 values for Z36 are close to the R1b frequencies, but the DYS437 and H4 values are not. A quick scan of the R-U152 database shows that these frequencies are also different from the other subclades of U152. And it just seems odd that its either 14/11 or 15/10. The percentages shouldn't fall that way.

Rathna
07-04-2013, 07:16 PM
The DYS 391 values for Z36 are close to the R1b frequencies, but the DYS437 and H4 values are not. A quick scan of the R-U152 database shows that these frequencies are also different from the other subclades of U152. And it just seems odd that its either 14/11 or 15/10. The percentages shouldn't fall that way.

That DYS391 is on the average percentage of the other haplogroups is due probably to the fact, as I said, that it rotates around these two values and the percentage is due by chance. The other markers have a different behaviour we may calculate just by the percentages you have found of mutations forwards and backwards. You should take in consideration also the fact, ascertained, that the same mutation rate may vary amongst different haplogroups or subclades. And don't forget that the haplotypes survived are only a few, the most part went extinct. You should read some postings that Jean Lohizun is posting on Rootsweb and I agree from so long with him (and I think he with me).