X matches

[Martha]

I am totally new to DNA and interpreting the results so my apologies if I mangle anything.

I have been looking at my X matches on a particular segment and categorising people who share the same or similar length and location.

My paternal half second cousin, Y and her two children, T & W [my half second cousins x 1 removed] match me on the same segment size and location however my paternal half second cousin, Y has a much smaller component at the same place than her 2 children.

I have looked three times to ensure I am not misinterpreting the results but it is consistently the same.

How and or why is this possible?

[georgian1950]

Are the children both female? If so they may have a more distant relationship via their father's line that is adding on to the length of the segment.

Are you using GEDmatch in this analysis, or a utility of the vendor such as the Chromosome Browser with Family Finder? With GEDmatch, sometimes you get into compatibility issues from the chipset used in the testing. However if you are using GEDmatch, look at the comparisons using the Graphics Bar. If you have streaks or bands of green for a full match, the segment probably is coming from being related more than one way.

Hope this helps.

Jack Wyatt

[Martha]

X Chromosome query

Hello Jack

The children are a boy and a girl. I am using the Chromosome Browser in Family Finder and down loading data to Excel.

Attached is the Excel report and the clipping of the diagrammatical representation of the data.

Martha

[georgian1950]

Hi Martha,

That is a bit of a puzzle. Do you have the kits on GEDmatch?

Jack

[Kathy Johnston]

Hello Jack

The children are a boy and a girl. I am using the Chromosome Browser in Family Finder and down loading data to Excel.

Attached is the Excel report and the clipping of the diagrammatical representation of the data.

Martha

Do me a favor and see if you can compare this mother with her two children. I do not match my own father in this same region due to version differences at FTDNA. If one tested a few years ago and another tested more recently they will not match each other in same location on the X. This has been reported to FTDNA but it would not hurt for others to report the same issue. Parents are expected to match their children along the entire length of a chromosome.

[Kathy Johnston]

Do me a favor and see if you can compare this mother with her two children. I do not match my own father in this same region due to version differences at FTDNA. If one tested a few years ago and another tested more recently they will not match each other in same location on the X. This has been reported to FTDNA but it would not hurt for others to report the same issue. Parents are expected to match their children along the entire length of a chromosome.

[Martha]

X Matches

Hi

Attached is a comparison between Martha, Mother, Daughter and Son when viewed from either the Mother, Daughter and Son's kit.

• Mothers Family Finder test completed 14 Mar 2013
• Daughter’s Family Finder test completed 5 Sept 2014
• Son’s Family Finder test completed 17 June 2013

Martha

[hansonrf]

Kathy, et al

In the region in question, 126198816 - 132770507, there are only 3.82-3.85 cM to distinguish. In my older datasets at ftDNA, they tested/reported on 3467 SNPs in this range; that was nearly 20% of all the SNPs [18091] they tested on the X. Interestingly, the data shown here has 3.85 cM match shown in this range [only 625 SNPs]; I suspect the new tests no longer 'waste' 3467 SNPs on such a fruitless region, leaving gaps where there is nothing to compare to. I personally don't have new data to compare myself...

It sure seems that the 'gap' is covered, though, with 3.85 cM in a range Rutgers thinks is only 3.82 cM wide...

Also, I don't think the algorithms know beforehand that you are comparing a parent/mother-child, so I'm not sure what they can do; maybe impute like 23andMe??...

Bob H.

[Kathy Johnston]

Kathy, et al

In the region in question, 126198816 - 132770507, there are only 3.82-3.85 cM to distinguish. In my older datasets at ftDNA, they tested/reported on 3467 SNPs in this range; that was nearly 20% of all the SNPs [18091] they tested on the X. Interestingly, the data shown here has 3.85 cM match shown in this range [only 625 SNPs]; I suspect the new tests no longer 'waste' 3467 SNPs on such a fruitless region, leaving gaps where there is nothing to compare to. I personally don't have new data to compare myself...

It sure seems that the 'gap' is covered, though, with 3.85 cM in a range Rutgers thinks is only 3.82 cM wide...

Also, I don't think the algorithms know beforehand that you are comparing a parent/mother-child, so I'm not sure what they can do; maybe impute like 23andMe??...

Bob H.

See this discussion a year ago:
http://forums.familytreedna.com/showthread.php?t=35434

[Kathy Johnston]

Hi

Attached is a comparison between Martha, Mother, Daughter and Son when viewed from either the Mother, Daughter and Son's kit.

• Mothers Family Finder test completed 14 Mar 2013
• Daughter’s Family Finder test completed 5 Sept 2014
• Son’s Family Finder test completed 17 June 2013

Martha

Yes just as I suspected. The Mother tested prior to April 2013. Most of the gaps on the X can be attributed to a nomenclature issue. The chips read the opposite strand within the double helix. Instead of reading A, T was read, Instead of reading G, C was read. Here is just an example using position numbers from Build 36:

122664306, rs11260444 AA now, was TT
125051065, rs5977829 GG now, was CC
126309782, rs5931945 CC now, was GG
126687359, rs5932136 CC now, was GG
129374997, rs7061846 TT now, was AA
132407803, rs5977874 AA now, was TT
etc. until at least rs12558861, enough differences to cause a mismatch when there should be a match.

GEDmatch has somehow fixed these gaps.
FTDNA has not corrected the gaps in matching.

Kathy

[prairielad]

another thing when looking at parent/child relationships and comparing to your own kit, you should set chromosome browser to show 1cM segments (default is 5cM), this will give you a better picture of matching.

Excel download's start stop points include these 1cM segments

Your chromosome browser picture of Martha to Mother(Y) does not include the 3.85cM segment that spans one these gaps produced by the different chip versions. I assume it is just showing 5cM and greater segments

Edit
If able compare the Build 36 X Raw Data between yourself and this Mother's (Y) between positions 123323504 to 126995150 and 132378403 to 134310817, You will probably see the situation that Kathy explains above, one will be C the other G or one will be A and the other T

CT and AG
TT and AA
CC and GG
AC and TG
etc.

[prairielad]

.........

Edit
If able compare the Build 36 X Raw Data between yourself and this Mother's (Y) between positions 123323504 to 126995150 and 132378403 to 134310817, You will probably see the situation that Kathy explains above, one will be C the other G or one will be A and the other T

CT and AG
TT and AA
CC and GG
AC and TG
etc.

these are part of the larger segment that shows Martha sharing more with the Son and Daughter then their mother

[Martha]

X Matches

Thank you Kathy. Your explanation makes perfect sense and it appears FamilyTree DNA have not fixed the issue.

Kind regards

Martha