Is Z36 in fact a mythological SNP?

[Kråke]

Is Z36 in fact a mythological SNP?
When I tested me for some years ago ( 23andme) my haplogroup was Z36. After that I tested me at FTNA and the haplogroup is Z36 - A7970 (A7967). Now I can see a discussion about that Z36 should not be included in the tree.
Thomas Krahn does not include Z36 in his U152 Superclade pack at YSEQ. He does allow you to test for the SNP however as an individual SNP but he does not recommend it. Thomas Krahn explained that the Z36 SNP is "99.1% identical to the X chromosome" and he does not consider it stable enough or have a proven phylogeny to include in the tree.

[Romilius]

Is Z36 in fact a mythological SNP?
When I tested me for some years ago ( 23andme) my haplogroup was Z36. After that I tested me at FTNA and the haplogroup is Z36 - A7970 (A7967). Now I can see a discussion about that Z36 should not be included in the tree.
Thomas Krahn does not include Z36 in his U152 Superclade pack at YSEQ. He does allow you to test for the SNP however as an individual SNP but he does not recommend it. Thomas Krahn explained that the Z36 SNP is "99.1% identical to the X chromosome" and he does not consider it stable enough or have a proven phylogeny to include in the tree.

I'm sorry, I'm not very into genetics... but... how is it possible that a SNP detected on the Y chromosome is someway disturbed by the X chromosome?

[theImmortal]

I'm sorry, I'm not very into genetics... but... how is it possible that a SNP detected on the Y chromosome is someway disturbed by the X chromosome?

I'm not an expert, but I've read about two ways in which this can happen. These may be the same process in some or all cases, but I recall that it is the second that Thomas Krahn is suggesting for Z36:

(1) The Y-chromosome and X-chromosome are apparently identical at the "tips" and actually do crossover in some way. This link shows where Z36 is on the Y-chromosome, though I don't know that this is close enough to the "tip" to be included as part of this region. (http://ybrowse.org/gb2/gbrowse/chrY/?name=Z36)

(2) There is a phenomenon called "X-Y gene conversion." The U152 subclade, R-BY3620 (on FTDNA, but known as R-L202 on YFull and R-P37 on The Big Tree), has a handful of SNPs that have been suggested to be the result of this process, and it appears to be good evidence that it occurs. (https://www.ncbi.nlm.nih.gov/pubmed/23810651). Basically, there are 4 SNPs (P37, P41, L202, and L303) that occur within a 14 base-pair region that have been found in kits within several distinct subclades (including R1a and L). The theory is that this is too coincidental to be random, parallel mutation, so the Y-chromosome must pick up some DNA from the X-chromosome.

I don't know if the researchers in this article found a segment of the X-chromosome that has these 4 mutations, but it sounds like Thomas Krahn has found a part of the X-chromosome that is basically the same as the Z36 region, and the SNP that constitutes Z36 is also found in the X-chromosome. This has been discussed here previously, and I believe the consensus is that Z36 serves as a reliable indicator of a parent subclade for many downstream U152 subclades, so it can be treated as such. Also worth noting, it does not appear that Z36 has been identified elsewhere, otherwise we would see it coined as Z36.1, Z36.2, etc., like P37 has been (http://ybrowse.org/gb2/gbrowse/chrY/?name=p37).

[gotten]

Z36 is in an X-degenerate piece of the Y-chromosome. Due to its origin (might be co-evolved from the common ancestor of both the X and Y chromosomes) it can strongly resemble some parts of the X-chromosome. YSEQ does Sanger-sequencing (also explained here), where you use a primer to pre-select a certain bit of DNA. If this region on the Y-chromosome strongly resembles a region on the X-chromosome you will get unreliable results. Sometimes the regions on the X-degenerate region are dissimilar enough to get a quasi-reliable signal, but apparently the region around Z36 is too similar to a certain part of the X-chromosome.

[R.Rocca]

Is Z36 in fact a mythological SNP?
When I tested me for some years ago ( 23andme) my haplogroup was Z36. After that I tested me at FTNA and the haplogroup is Z36 - A7970 (A7967). Now I can see a discussion about that Z36 should not be included in the tree.
Thomas Krahn does not include Z36 in his U152 Superclade pack at YSEQ. He does allow you to test for the SNP however as an individual SNP but he does not recommend it. Thomas Krahn explained that the Z36 SNP is "99.1% identical to the X chromosome" and he does not consider it stable enough or have a proven phylogeny to include in the tree.

YSEQ uses Sanger sequencing to detect SNPs. When creating primers to call a region of the Y-chromosome, the primer might call something similar/identical in the X-chromosome instead, thereby producing a false positive. That is why it can't be used reliably when building a tree based on Sanger sequencing. However, using whole genome sequencing (like Full Genomes or BigY), one does not need primers. In this case, detecting Z36 is no problem at all. Trust me, Z36 is a real SNP.

[Romilius]

Thank you for answers. So, we have to consider reliable Z36 subclades SNPs? For example A7992, A7993 and others?