Where did the phased father's X come from?

+5 votes
A son and mother have GEDmatch IDs where the source for both was FTDNA.

A phased father (and as it turns out phased mother) were created from the son's and mother's results.

The phased father has a 4th cousin one-to-one match to the mother aka his wife. Expected actually.  But the phased father also has a X match to the mother (to original DNA kit and phased M1 kit) aka his wife. Not expected.

If the mother's DNA is "subtracted" from the son's to get the father's phasing, where might the phased father's X have come from?

The son as a YX got the Y from his father and the X from his mother so there was no X contribution from the father to the son so...

I feel as if I am missing something simple.


in The Tree House by Pete Toemmes G2G4 (4.8k points)
This seems like just a mistake by GEDmatch.  I've never heard of any X chromosome information being passed by the sperm, so the father's X chromosome is completely inaccessible.  Any relationship between the father and mother is completely irrelevant, since the phasing only deals with the sons and mothers DNA.

The only other possibility is if the "son" is actually a female, or a daughter was included.  A daughter would have received the entire father's X.  I assume you are certain the son is a male, and no daughters were included?
He is a male and he does Y-DNA match other males on FTDNA and the atDNA matches on GEDmatch support maleness so I do suspect a hiccup in the kit transfer from downloaded FTDNA into GEDmatch.




Pete, after reading your response to Ken about the size of the X match to the phased father, and Rob's post above, a silly idea occurred to me. When profiles are created at GEDmatch, we specify whether the kit is for a male or a female; this is also editable, of course, by selecting the profile/kit under the "Edit or Delete" option from GEDmatch's main screen.

Is there any earthly possibility that your cousin's GEDmatch profile is set to female rather than male?

I am fairly certain the male/female indicator on Gedmatch is only informational and doesn't affect the results.

I am also fairly certain that the son's raw data contains 2 values for the x-chromosome SNPs. It would be the only way the paternal phased kit could have any x-chromosome values when phasing a mother and son.

One of the most repeated warnings about keeping DNA information private is that you might learn something about yourself or someone else that is uncomfortable. This seems like a prime example.

IMO, there is a real possibility that someone may have an undiagnosed medical condition and this person is both identifiable and contactable. I would want to know.   

I would contact this person with my concerns.  I might even suggest the person delete the existing Gedmatch kit, remove the duplicate x-chromosome SNP value from the raw datafile, and then reload the kit. The results will be more accurate and some privacy can be retained.

edit: I noticed Edison tested on AncestryDNA which has 2 SNP values for the x-chromosome which may be the answer.  23andme and FTDNA appear to report only 1 SNP value.
I wanted to correct the statement that FTDNA contained only 1 value for the x-chromosome. I looked at the kit that I had uploaded from 23andme, it has 1 value. The FTDNA typed kit has 2 values.
And in my learning stuff every day category, based on your post about positional reporting on the X I looked at two sets (different males) of raw data from AncestryDNA v2 tests, and they both show 17,605 SNPs ranging from rs5939319 at 2,700,157, to rs669237 at 154,916,845. The positions are based on GRCh37 mapping, but should be relatively close for both Ancestry and GEDmatch.

It looks like the OmniExpress chip skipped the largest of the pseudoautosomal regions, which should end at about 2.6 Mbp...and it's likely it's skipping the tiny PAR2 region at the end of the long arm of the chromosome, too. But otherwise, the sampling looks pretty much along the entire length. I'd be interested to know if 23andMe and/or FTDNA sample lower than about 2.7 Mbp. Also curious how the GSA chip performs with respect to this.

Also 884 SNPs on the Y-chromosome in there, and 439 mtDNA SNPs. I knew some were sampled, I just never bothered to look closely or count. Project for the weekend may be to see if I can find a mapping of the Y-SNP reference clusters to any corresponding alignment with those cataloged for haplogrouping.

Okay. Sorry, Pete. Not gonna drive this farther off topic since we think we have the answer to your question. Just something that made me go, "Hmm."
Oh - not at all.  Digress away.  This is very interesting stuff especially for me as it expands my ever growing knowledge base from a low(er) baseline.

It is my goal to become a stable genius like...sorry... for the late Friday "humor".


In response to your query about ftDNA showing anything below 2.7mBP I've checked my collection of donors and uniformly they show the same as ancestry, except for one single SNP at rs17883004 at 1.4mBP. That is reported on 4 out of 7 samples - the others being no read for that position.
Thanks for checking, Derrick! I suspected as much. After I saw your note, I glanced at the Illumina product spec sheet for the "Infinium OmniExpress-24 v1.2 BeadChip" and it lists 17,565 markers tested on the X-chromosome (my raw Ancestry data is consistent, but shows 17,605) with 683 tested at PAR (pseudoautosomal) loci. A little more digging indicates that, for the build 37 map, the OmniExpress-24 considers 62,321 to 2,697,868, and 154,939,018 to 155,236,747 to be the PAR regions. The same digging also indicates that, for genealogy testing, the companies routinely discard the PAR data even thought the chip supplies it; understandable: that potentially heterozygous data doesn't have any genealogical bearing and would just muddy the waters. But it's also unsurprising that a PAR marker or two will slip through into the published results with some frequency.

On the yDNA front, I also suspected my inclination to check the Y-SNPs shown with the autosomal results wasn't very original. And it wasn't.  :-)  Our own Dr. Turner even commented on it a couple of years ago.

Net message is that those tag-along Y-SNPs that are captured in atDNA tests really don't tell us anything. An FTDNA project manager compared the AncestryDNA SNPs to those in the ISOGG index and found that of the 885 SNPs he examined (the total shown in his AncestryDNA results), only 318 were in ISOGG's index. The remainder did list in the Ybrowse database (but that uses an NCBI full genomic dataset, so I'd fully expect all the reference clusters to appear there, just as they do in the "1000 Genomes Browser"). And of the 318, 256 can be listed as "relevant" to a haplogroup because they appear in the mix somewhere...but in general they're too far upstream to tell us anything without having the full complement of hierarchical SNP results. For example, he found 79 SNPs related to haplogroup E, 72 to R, 40 to I, and the rest had 19 or fewer SNPs.

Too, there seem to be issues with some heterozygous calls and false positive results in these atDNA Y-SNPs...I suppose understandable since the tests aren't for-purpose and don't go through the multiple sampling and quality control that a Y-SNP test would. So even if the SNPs tested were more directly applicable to yDNA haplogroup analysis, there would be questions about accuracy and consistency in doing so. On the bright side, though, that means I finished my weekend project by noon Saturday.  ;-)

P.S. I'm done now, Pete. Honest.
Yeah, but USA men's curling won the Gold first.

Curling is almost as hard as DNA analysis - certainly a little bit more physical ;-)

3 Answers

+3 votes
Best answer
Some companies actually provide homozygous genotypes for the X in males. GEDmatch doesn't have any exception handling in place for phasing a single X. Just ignore the X chromosome for P1 kits.
selected by Pete Toemmes

Ah. Makes more sense now. And it made me realize that I've never done solo-parent phasing at GEDmatch before using a mother and a male child. I looked at my raw data from AncestryDNA and, sure enough, my hemizygous X appears as homozygous, with a pair of duplicate alleles at each reference cluster (17,605 SNPs; I checked just for grins). So it looks like John Walden's algorithm works on exactly what data it's provided. Learn something new every day.

Many thanks, Ann!

Edited to add: Since it's Pete's question, I don't want to mark Ann's as "Best Answer," but I do think she cleared the whole thing up in about 30 words.  :-)

Done.  Not that any of the other responses aren't of high quality - they all are.  My cousin is following this thread and will comment if he feels appropriate.

Glad this likely uncovered some new info for those of you who are much further along that I - and that is most of you which is why I asked ;-)

Yes - privacy is of paramount import in all of this and the concerns expressed are very much appreciated.



+4 votes
Is the  father a "blood" relative of the mother?  A maternal grandmother, for example.  He could have inherited it from his mother who was descended from same grandmother as his wife's.  I know that's endogramy (inbreeding), but X a woman's X chromosomes get passed to her daughters as well as her sons.  Grandmother to daughters to their offspring (male or female) and finally to the son.
by David Hughey G2G Astronaut (1.3m points)
Thanks.  They are most certainly not that close in blood relationship.  In fact it's a bit complicated and that which we hope to sort out.  

We are almost certain that they are from two family lines that share an unidentified common paternal ancestor at MRCA 3.5 but there is a bit of endogamy at play.

The son has a paternal line as well as a second input from the same line via a female of same maiden surname at his paternal GF level (this person's mother) so I can see that an X inherits to the paternal GF but not to his son, who is the father of the son of interest. The target son's mother descends from a male line of the same surname that we hope to prove is part of the common male ancestor.

Still very strange to me. If this is not a GEDmatch phasing glitch then there is perhaps something amiss in the family tree(s) we believe to be accurate.
You should be looking for the same mother as a mother's X-chromosome passes to both her male and female children. It's your father's mother's X-chromosome that is in question.
+2 votes
According to Gedmatch, I share a small 6.3 segment, 397 with my father, which should not happen.  This seems to be caused by the DNA Service reporting 2 strands of DNA at the beginning of the X-chromosome.  

You should be able to verify for yourself.  This appears in my gedmatch 23andme data. I don't know if all DNA Services report the beginning of the x-chromosome. I read this is normal, or at least not unusual.

Theoretically, the phased father, in this case, would not have any x-chromosome, but might for this particular region. I believe that Gedmatch will report a shared segment on the x-chromosome between males for 1 or 2cM and 200 SNP's.
by Ken Sargent G2G6 Mach 5 (56.7k points)
Interesting.  All three were born in the same geographic region and if our theory of a shared unidentified paternal ancestor holds true then, geographically, this more recently split line traces many many generations from the same general geography.  FWIW, the paternal haplogroup is J-M267/J-L174.1 non Jewish (religious) but Cohen/Kohanim ethnic.

The phased father and son do share one 190.1 cM X.

The phased father and mother (his wife) share the same (location) 190.1 cM X.

The son and mother hare the same (location) 190.1 cM X.

I suppose if we had a true DNA sample from the father, who is deceased, we would likely not see a match.  I'll have to try to find someone who matches that criteria and give it a whirl.



One of the most significant and volatile issues on 23andme centered around Privacy.  It got so bad, after years of participating, I and many others stopped participating because every post turned into a privacy debate.  

This post would certainly have generated a lot of interest.

Given that most people on 23andme had, at least at one time, subscribed because of health, traits, or Ancestry Composition, the response to your post would probably include the following. 

The son may have Klinefelter syndrome. This particular phased Paternal kit should theoretically have no x-chromosome SNP data at all, and therefore share no segments with anyone.

I don't know enough about the subject, so I would be cautious. Also, I can't say with any degree of certainty, that some technical problem with gedmatch phased kits, but in my own test, I didn't share any other segment with my father other than what I already posted.

Klinefelter syndrome is a common genetic condition affecting males, and it often isn't diagnosed until adulthood.


This would be the type of story that gets's headlines if true.

edit: the x-chromosome is about 196cM so the shared segment described is the entire x-chromosome.

I believe it to be a glitch.  I may report it to GEDmatch as they might be interested.  Yeh - 190.1 cM is a big red glitch flag.  Might also get him to regenerate the phased father too.

He has fathered 7 children and, though I've never met him personally, his FB family pic suggests he's as normal in this regard as any father.

He's a cousin of mine.

I appreciate the link to Mayo.  Quite interesting.


Related questions

+26 votes
2 answers
+6 votes
2 answers
277 views asked Nov 12, 2017 in Genealogy Help by Barry Smith G2G6 Pilot (159k points)
+6 votes
1 answer
122 views asked Jan 15, 2018 in The Tree House by Emma MacBeath G2G6 Pilot (693k points)
+11 votes
2 answers
361 views asked Dec 24, 2017 in Genealogy Help by Karen Raichle G2G6 Mach 7 (73.1k points)
+4 votes
3 answers
411 views asked Jan 4, 2018 in The Tree House by Stephanie Stults G2G6 Mach 3 (37.0k points)

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright