Family Tree DNA Forums

Family Tree DNA Forums (http://forums.familytreedna.com/index.php)
-   X-DNA Advanced (http://forums.familytreedna.com/forumdisplay.php?f=233)
-   -   Understanding x-match (http://forums.familytreedna.com/showthread.php?t=42507)

alice 29th October 2017 03:03 PM

Understanding x-match
 
I am baffled by the match described below. Please, can someone with a better understanding of the x-chromosome help me?

F and M are the father and mother, respectively, of Daughter 1 (D1) and Daughter 2 (D2).

F and his children have a fairly strong autosomal match (38 cM in one segment) with Cousin (C).

M and C have no autosomal or x match.

F and C have no x match.

From triangulation it is clear that the match between F and C is on F's father's side, not his mother's.

Yet, D1 and D2 both show x matches of the same 7.9 cM segment with C. How is this possible, since the x that F passed on to his children was from his mother?

Thanks a lot.

prairielad 29th October 2017 04:27 PM

Depending on size of X match, ignore it as it maybe false.
It is recommended that only X matches over 10cM will be valid.

FTDNA declares anyone a X match if segment is 1cM/500SNPs or greater.

Women to women X comparisons will have more false X matches(they each have two X values) then a male to male comparison (they only have one X value each). Algorithm is only selecting 1 value to match in a single sequence of values to create a matching segment.

prairielad 29th October 2017 04:33 PM

Phasing parents to child at Gedmatch may show that X match is not valid if this match also utilizes Gedmatch.

alice 31st October 2017 10:19 PM

X-comparison on Gedmatch
 
I tried comparing daughters' phased paternal kits with C's kit but got an error message: "both kits must have an x-chromosome." I don't understand because of course both do have an x-chromosome.

ltd-jean-pull 31st October 2017 11:03 PM

If C is not an X-match to F nor M, but F&M's daughters are an X-match then is very likely to be due to them being "identical by chance" on the X-chromosome.

There is a very good description of Identical By Chance in this article by Roberta Estes.
https://dna-explained.com/2016/03/10...on-and-chance/

If you have a look at a male kit and a female kit on Gedmatch the females have several-fold more X-matches than the males. Because they have two X-chromosomes and males have one you'd expect them to have only twice as many X-matches as a male, so many of these X-matches can't be real.

prairielad 1st November 2017 01:31 AM

Quote:

Originally Posted by alice (Post 445025)
I tried comparing daughters' phased paternal kits with C's kit but got an error message: "both kits must have an x-chromosome." I don't understand because of course both do have an x-chromosome.

How long ago were these kits uploaded to Gedmatch, Original FTDNA uploads required one to upload 2 files, autosomal Raw Data and X Raw Data. X Raw Data was optional.
Now one has to upload the concatenated Raw Data file from FTDNA which has X included.

Run Gedmatch diagnostic utility on kit numbers in question and see if X (chromosome 23) shows a Number of SNPs used.

dna 1st November 2017 03:02 AM

Quote:

Originally Posted by ltd-jean-pull (Post 445026)
If C is not an X-match to F nor M, but F&M's daughters are an X-match then is very likely to be due to them being "identical by chance" on the X-chromosome.

There is a very good description of Identical By Chance in this article by Roberta Estes.
https://dna-explained.com/2016/03/10...on-and-chance/

If you have a look at a male kit and a female kit on Gedmatch the females have several-fold more X-matches than the males. Because they have two X-chromosomes and males have one you'd expect them to have only twice as many X-matches as a male, so many of these X-matches can't be real.

Yes, Identical By Chance.

Yes, females have two X chromosomes.

However, you also need to take into account that females are related to more people in their family, than males are. Please take a look the following two pictures that show how very different X chromosome inheritance patterns for females and males are:

http://genie1.com.au/images/blog-pic...s-LCoakley.png

http://genie1.com.au/images/blog-pic...s-LCoakley.png

The above images came from the page http://www.genie1.com.au/blog/63-x-dna


Mr. W

alice 2nd November 2017 11:31 AM

X-match and phasing
 
Thanks to everyone who replied to my question; you have been very helpful. My latest steps may be helpful to others as well.

FTDNA shows no X-match between F and C, but both daughters show X-matches with C. (X-match of 3 cM on FTDNA and 7 on Gedmatch).

On Gedmatch I checked using the diagnostic utility and it showed F's X-chromosome having only about 10,000 SNPs. I deleted that kit and re-uploaded the raw data file and F's X-chromosome went up to 19,000 SNPs.

I then re-ran the daughters' phased files and compared the phased X-chromosomes with C's. No match. However, their unphased files do show the 7 cM match (though there is no X-match with M, their mother).

So my conclusion is that this is all much ado about nothing and I am confirmed in my original assessment of where the connection between F and C lies. I wish there were a way for FTDNA to do some of this legwork for us and not show the match in the first place!:)


All times are GMT -5. The time now is 06:50 AM.

Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.