Interpreting X DNA?

Question

I guess I need a "dumbed down" explanation of these things, since I don't understand the explanations I've read.

First, I have a 3rd cousin match from Ancestry who uploaded his info for me on gedmatch.com, and I see that he shares no X DNA with me. We're trying to figure out which of his great great grandparents we share (I'm searching for my biological family). Initially, I thought this simply meant that we wouldn't be related through my bio mom's family or his mom's family, so I was going to dissect his father's side of his tree. After researching a little more, I'm thinking it's not as simple as that. I only got a "C" in Statistics while working on my Masters, if that helps you understand how simply I need this explained;)

Additionally, I have a cousin match on ged match who shares 77.3 cMs worth of X DNA with me and no autosomal DNA. How much of a cousin match is this? There is nothing in the Gen column for this match, but I don't know why.

Answer 1

One of my favorite math teachers in high school told us that whenever we got stumped, we should draw a picture to get a better definition of what the problem was.

Here's a recent (12 June 2015) article by Louise Coakley, X-DNA's helpful inheritance patterns. Coakley published several graphs with the possible lines of descent marked out.

Coakley says:

With no X inherited from some ancestors, varying amounts inherited from others, the randomness of DNA recombination at each generation, and occasional sticky segments passed down intact over several generations, X-DNA can be quite unpredictable and difficult to interpret exactly where it came from.

Before you try to sort out where the match could have come from, take a moment to think about how the X gets passed down for males vs. females.

Coakley says:

Because males only inherit an X chromosome from their mothers, if a male has an X-match in his DNA results, the shared ancestor must be on an ancestral line that follows the male X-inheritance pattern

and

Because females inherit X chromosomes from both parents, if a female has an X-match in her DNA results, the shared ancestor must be on an ancestral line that follows the female X-inheritance pattern

Below each of these statements she has a chart where the pattern is marked out in green. That shows each of you where your own Xs could have come from. It is not easy to work any of these problems out when you are only comparing a match between two people, and it is easier to work out where the X could have come from if you had a match on it.

For your second case, with the cousin who has an X-DNA match but no autosomal, Coakley says:

It is common to share segments of X-DNA with people who share no significant amount of autosomal DNA.

The situation is complicated further because men inherit the X from their mothers unchanged, but with women, they get an X from their mom which has DNA recombined from both their mom's X chromosomes. If this gives you a headache just to think about it, you aren't alone.

Coakley has a section in the article Practical uses of X-DNA where she goes over different scenarios and how the X-DNA might pass down.

She has links to download the entire article in PDF format, high-resolution versions of the charts, Excel templates to help you map out matches, and more links to other articles that talk about X-DNA.

Sometimes the explanation of a topic that makes it clear for one group of students doesn't work for everyone. Instead of recognizing that the teacher hasn't made the topic clear yet, the students who didn't "get it" on the first try internalize this mis-match and say it is because they are dumb.

The real issue is that some subjects are difficult, and we aren't always fortunate enough to discover the explanation that works the first time we try. The students who are forced to take a second and third look often end up with a better understanding of the topic in the long run, because they've had to think about it more. If this article doesn't make it clear, follow her other links and read more articles. Give yourself some time to let things sink in, and don't be intimidated if you have to read articles multiple times before you feel like you 'get it'. My experience is that even when I think I've understood something the first time out, it can take six months or more before the information really sinks in.

Answer 2

The oldest answer has given a great overview of X inheritance, and I agree with what it says. It does take study to understand. Initially I suggest you mainly rely on Coakley's two main charts of X inheritance (one for a female descendant, one for a male descendant). The green boxes on those pedigrees show who could have carried the X segment you are concerned with.

That said, what about your first specific situation? What about your "third cousin" who shares no X with you? This fact gives you no help in determining which of his or your ancestors could turn out to be your common ancestor. The lack of X coming to you does not imply anything about the male & female connections involved. This is because it could have been a from a potential-X-contributor but the randomness of inheritance simply did not let any pieces of that ancestor's X come down to both of you -- just random luck. So don't try to narrow the search for this case.

Your other cousin shares 77 cM of X, though, and in this case you can apply the charts given by Coakley to limit which ancestors need to be considered. The presence of the X DNA in both of you shows that the path had to be through the green boxes. (Remember to use the female chart for your paths back to candidate ancestors, and the appropriate chart, male or female, for the cousin's paths back.)

The number of generations back is complicated for X inheritance: the likelihood varies between each of the possible paths depending on how many males are on the particular path. The fewer males involved, the fewer generations back are probable; the more males, more generations are likely. (This is because the X does not get divided into pieces when going from father to daughter, so has a "free ride" for that generation.) So, X generations are not simply characterized, and that is probably why GEDmatch omits giving a "generations back" estimate. (Remember, the Gen column is only an estimate in any case, and should not be taken as being precise; in fact, it can be a fair number of generations too low for small match amounts.)

Answer 3

You said: "First, I have a 3rd cousin match from Ancestry who uploaded his info for me on gedmatch.com, and I see that he shares no X DNA with me. We're trying to figure out which of his great great grandparents we share"

Your cousin is male. Males only get one X and it's from their mother. The simple way to think of this is that the X cannot be passed down to two males in a row.

Therefore of his 16 great great grandparents, since you don't share X, you could be related through any of the 8 great great grandparents on his father's side, i.e.:

• his father's father's father's father
• his father's father's father's mother
• his father's father's mother's father
• his father's father's mother's mother
• his father's mother's father's father
• his father's mother's father's mother
• his father's mother's mother's father
• his father's mother's mother's mother

(I've bolded the "2 males in a row")

or through these great great grandparents on his mother's side

• his mother's father's father's father
• his mother's father's father's mother
• his mother's mother's father's father

The 5 great great grandparents you likely are not related to are:

• his mother's father's mother's father
• his mother's father's mother's mother
• his mother's mother's father's mother
• his mother's mother's mother's father
• his mother's mother's mother's mother

So X doesn't help you very much, as only 5 of the 16 candidate ancestors are not likely.

The reason why I say "likely" is that a woman gets the one X chromosome her father had, but only one of the two X chromosomes that her mother had. It is therefore possible that for one of the 5 great grandparents listed just above, that the common ancestor's segments did not get transmitted down mother to child. I'm not sure of the exact probability here, but at 3rd cousins, it may happen but not too often. This means none of the 16 great-grandparents can be assuredly omitted. So unfortunately, the bad news is that the X chromosome is likely not to be of too much help to you.

Instead, I would recommend you look at the Autosomal DNA and the people you and your 3rd cousin both match with on a particular segment. This requires chromosome matching, which you can do at gedmatch. The great great grandparent you are looking for may be able to be determined from the common ancestors of you, your 3rd cousin, and the people you both match with.

---

Then you asked: "Additionally, I have a cousin match on ged match who shares 77.3 cMs worth of X DNA with me and no autosomal DNA. How much of a cousin match is this?"

The X chromosome does not crossover as often as other chromosomes. Therefore a match on the X chromosome may continue many more generations than it does on the other chromosomes. So it is quite possible that a 5th or further cousin may share on the X but not on any of the other chromosomes.