How much X chromosome is a DNA match for confirmation?

Question

The directions page does not give a threshold.  Please edit that page to list the cM threshold before an X chromosome match is considered enough for DNA confirmation.  Is it 7 cM too, or another threshold?

Answer 1

7 cM if a female is included.  3 cM if the match is between only two males.

Comments

Peter, this notion of a 3-cM segment being sufficient for a male doesn't make logical sense. Please see Helen Ford's comment at https://www.wikitree.com/g2g/400173/are-x-dna-ancestor-percentages-misleading. If Helen and her son (who got all of his X DNA from Helen) both share (say) a 4-cM segment of chromosome X with a cousin, what is the logic that says the son can claim DNA confirmation based on the segment, but Helen cannot?

---

Thanks Ellen,

I should have also added the amount of shared X-DNA needs to be sufficient for the relationship.

Sincerely,

---

I have tried many times over that past few years to try and explain just how logically incompatible the guidelines are related to auDNA.

They are opposite to what I have been taught.

Triangulation is always used as part of a process that involves labeling segments.  The first step is to identify all positive matches. These are individuals who BOTH have known ancestral lines that intersect and DNA match.

A DNA match requires least 1 IBD segment which is not IBS or IBC.

Here are some of the things that Wikitree does the opposite of what I have been taught.

1. Even though only DNA matches that share at 1 IBD segment are used, Both IBD and IBS are used and tracked.  IBS is considered gold or at least silver but the opposite is true on Wikitree, and they are considered poison.

The first step ONLY involves known relationships in which of these IBD and IBS segments are labeled which at least identifies the line as on the paternal side or the maternal side. Wikitree draws the line at 3rd cousins, but there is no such line outside of Wikitree.

The X chromosome is different.  It is not used to match, but it is used on 23andme as part of the relationship prediction. I don't believe any other service uses the X-chromosome, but they would report it.

Only Wikitree uses x-chromosomes to match, and only Wikitree believes that male to male is a special matching case.

Wikitree uses the total segment shared as part of is validation, but It's not logical, at least to me, given the total segment lengths include IBS segments to use the total segment length as evidence in this case but consider them poison.

---

Here is an example I think I have used in the past.

An ancestral couple beyond 3rd cousins have all daughters, and those daughters only have daughters.  2 female descendants are tested and match on the x-chromosome but their tests can't be used.  But they can be used if both females have at least one male child that also shares a segment x-chromosome. This doesn't make sense.

Answer 2

I have found somewhat contradictory information on various typically good sites...

Special Treatment for X Matches  from 

https://dna-explained.com/category/x-chromosome/

While the generally accepted threshold for autosomal DNA is about 7cM, for X DNA, there appears to be a much higher incidence of false matches at higher levels than the rest of the chromosomes, as documented by Philip Gammon as in his Match-Maker-Breaker tool.  This appears to have to do with SNP density.

I would encourage genetic genealogists to consider someplace between 10 and 15 cM as an acceptable threshold for an X chromosome match. This of course does not mean that smaller segment matching can’t be relevant, it’s just that X matches are less likely to be relevant at levels below 10-15 cM than the rest of the chromosomes.

But the entire article is written as looking at X as an enhancement to typical matching on the 22 chromosomes.  So it is inferred but not explicitly stated that you have to first match at 7cm or above to a chromosome.  

http://blog.kittycooper.com/2014/01/what-does-shared-x-dna-really-mean/

robertajesteson May 9, 2016 at 3:42 pm said:

Family Tree DNA only shows X matches if there are other matches as well over the match threshold. X matching alone, especially without other matches, can be deceptively high and from very far back in time, which does not really play by the same rules as the rest of the autosomes. I’m assuming that is why.

robertajesteson June 4, 2015 at 8:57 pm said:

I would rely on other information from other chromosomes primarily. I have found the transmission characteristics of the X to be somewhat different than the rest of the chromosomes so I could be hesitant to apply the same guidelines to the X as to the other autosomes. I don’t believe Blaine Bettinger included the X in his cM study, but it would be very interesting if he were to do so.

robertajesteson June 3, 2017 at 10:01 pm said:

Keep in mind that X chromosome fragments generally speaking need to be about twice as large (cM) as other chromosomes to be a valid match.

http://www.yourgeneticgenealogist.com/2014/01/ftdna-releases-x-chromosome-matching.html

From http://smithplanet.com/stuff/gedmatch.htm:  

• X-DNA is different. The cM match amounts are not very helpful in determining how closely related you are.
• The X Chromosome is not passed from father to son, so the lines between you and a strong (over perhaps 4cM) X-DNA match and your common ancestor will not have any father-son relationships. You'll notice in the screenshot above that I (being male), as expected, have 0cM X-DNA match with my paternal grandmother and 69.7cM match with my maternal grandmother.
• If you are a male, any X-DNA matches will be related to you on your mother's line. If your X-DNA match is male, you will be related to him via his mother.

You can read more about X Chromosome matching at http://smithplanet.com/stuff/x-chromosome.htm.