GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-11 > 1289337502


From: Kathy Johnston <>
Subject: Re: [DNA] X-Y "recombination" (was Re: WTY Update slides)
Date: Tue, 09 Nov 2010 16:18:22 -0500


Hi Bonnie,
I hope you saw my e-mail to you warning you of my note on this list so you would not be too surprised. Thanks for responding!

I took Thomas literally when he said that the J2a4i founder had a SNP from the X so I assumed it came from the “C” at rs12859783. I figured this SNP along with surrounding SNPs were transferred directly from the X to the Y during ancient times. So do you or he think it could have come from a different part of the X or somewhere else? Maybe this idea is not widely accepted by the academic community, but how would we know if nobody ever publishes it and asks the question?

The only reason I used the term phasing is because at HapMap, as I recall, almost all the females in the CEU European group can be phased by using the three haplotypes represented in the Google Doc spread sheet. Any time the females are consistently a combination of two male haplotypes, I think of it the same as phasing, but more straightforward, so maybe I should just call it haplotype mapping. HapMap also allows for looking at parent child trios but these are really not needed in this location because we (including most of us genealogists) are easily haplotyped. I sent you and Thomas an example a few months ago of the large group of SNPs surrounding the rs12859783 provided by HapMap. I thought these represented three distinct, but very ancient founder groups. I also thought that these were so old, that you probably could not connect these with any specific geographical area. I do remember there was clearly an association with Europeans. The hypothesis was that if your father’s ancestor was the first J2a4i, then his father’s Y came into contact with an X from his mother (paternal grandmother to the J2a4i founder) that presumably carried the Haplotype C (featured on our Google Doc spreadsheet).These X and Y haplogroups would have co-inhabited the same locale during the same time period. And why is our Haplotype A so drastically different? Did one of these come from Neandertal? LOL. These are all pretty common haplotypes.

This kind of research may be way beyond the capability of a bunch of us population biology wannabees. We need to ask people involved with hardcore sequencing and population biology I would think.

When I said redacted, I meant that the SNP we want to know about has been removed (or never added) from Illumina and Affimetrix Chips used in DTC testing, so we need to go to HapMap to find the one that is missing. It seems pretty easy if you accept the idea that there are specific haplotypes present and the sequences hold up here over time.

If there are other SNPs on the X that have rs names involved in the block that was transferred, I would sure like to know the names of the other ones too. Even if the transfer came from a different location on the X, the technique is really the same, what we should call the “Joy of HapMapping” for those of us who don’t like to cook.

IF your father had been tested for rs12859783 by the HapMap group, would not the chip have picked up a false positive heterozygous result? Maybe not if the flanking sequences are distinct. 23andMe used to show some heterozygous results for males in the raw data,then changed all these to no-calls when it created a stir. There has never been a test for the rs12859783 by our testing companies, but I assumed that an X to Y transfer could also create a false positive test if it had been available. A heterozygous result makes the SNP appear like it was happening in a female. Does anyone think any of those heterozygous SNPs originally reported by 23andMe jumped ship from the X to the Y?

Now that you have found all those BLAST similarities, it does throw a monkey wrench in all of our efforts, but until you find another C at rs12859783, I can at least dream that one of my ancient X cousins was hanging out with the J2a4 Y cavemen.
Kathy J.



This thread: