X phasing project - opinions/comments please

[Canyon Wolf]

I am trying to phase the x chromosomes for 3 siblings and would appreciate another set of eyes to look things over.

I have a male (T) and 2 female (S & K) siblings. My interest is with the maternal Xs of the mother.

1. T (male) has only 1 X and it must be maternal (but could be a mix of Mom's 2 Xs)

2. S & K (females) will have an identical X (except possibly the very ends) which came from their Dad. Therefore they will show a 1/2 identical match for the full length of their X chromosome. Therefore, I should be able to ignore this paternal contribution for the purpose of this exercise.

3. The 2nd X of S & K is maternal and can be a mix of their Mom's two Xs

4. The mother has two Xs, one came entirely from her dad (who got the whole chromosome from his mother) and the other came from her mom (which can be a mix of the two maternal grandparents).

X Chromosome Map Link

After determining the cross-over locations and identifying their owners, I mapped out the maternal X contributions for each sibling (see the jpeg).

I started with the male who shows 2 cross-overs. I arbitrarily assigned maternal grandparent 1 (M1) to the two end segments and maternal grandparent 2 (M2) to the middle segment. I then used the matching segments between the 3 siblings to map which female received segments from which maternal grandparent.

It appears that K received an un-crossed X from maternal grandparent M1. If this is true, then my next step is figuring out which maternal grandparent is M1. ;-)

Would anyone care to comment? PLEASE

Did I forget to take anything into account? Have I made errors in assumptions? Does anyone think there could be additional cross-overs hiding in there?

[Canyon Wolf]

Additional Thoughts

I have kits for 3 male siblings of the mother (K's uncles) and a daughter of a 4th male sibling. These kits represent male X chromosomes, therefore can only match with K on her mother's maternal X.

> K does not match any of her uncles or the cousin on the X.

I also have the kit for a female sibling of the mother (K's aunt). K is a match for the full length of the X with this aunt (FTDNA chromosome browser). At gedmatch they are 1/2 match for full length of the chromosome with widely scattered very small full matches spread over the entire length of the X chromosome.

The father of T, S, & K and their mother's father are both colonial New England stock so this may account for these little full matches - I think - identical by population.

I think this additional data strengthens my hypothesis of no cross-over for K and suggests the X that K received probably belongs to her mother's father.

[Kathy Johnston]

I think your assumptions are probably correct. T could have a barely noticeable crossover at the very end of the X chromosome but this won't give you any grief because it is way too small to be significant.

[Canyon Wolf]

Thanks - I really think I am starting to get a good handle on phasing. I guess it is time to go back to the autosomals and see how far I get phasing them before I run into trouble

[Kathy Johnston]

Thanks - I really think I am starting to get a good handle on phasing. I guess it is time to go back to the autosomals and see how far I get phasing them before I run into trouble

There are some strategies that you learn if you practice. If you are just starting out, pick an autosome that has some non-identical and some fully identical regions between sibs. Pick one that has very few crossovers for at least one member of the family.

I use what I call a skeleton of crossovers that I take from GEDmatch. In your example above, for the X, it ends up looking like your picture. If someone has no crossovers (like K), then you simply colorize the entire chromosome because there are no borders to stop at. Two chromosomes are tricky but often can be done if you start with what you know first before making any guesses.

[Canyon Wolf]

Thanks again Kathy.

X Phasing this group of 2nd generation descendants has been extremely useful to my project.

You have been so helpful concerning this phasing stuff. You got me started with a 1st generation autosome map not too long ago but things went down hill. So I decided to try the Xs. First thing I discovered using the gedmatch graphics like you showed me was a crossover in the 1st generation group that was not being picked up by FTDNA or gedmatch using the default settings. The matching area was less than 5cM! But it showed up clearly when I stacked the siblings up.

I think the hardest part for me when I tried the Autosomes was dealing with the fact siblings can and will match each other twice in many places. When I started I only had 4 1st generation siblings and the phasing maps looked pretty good to me. Then I had a close cousin to work with. I just keep going around in circles when I try to add the cousin into the map to determine which chromosome came from which side of the family. I must have missed something because things just won't work out right.

Now I have a number of descendants from additional 1st generation siblings (with more coming in thanks to the xmas sale and coupons - boy don't I love those coupons ;-). Since the new kits can only match the 1st generation siblings one way I may have a little better luck keeping things straight in my head. Or NOT! Only time will tell, but I am determined to try.

Do you have any additional words of wisdom?

[Kathy Johnston]

Thanks again Kathy.

X Phasing this group of 2nd generation descendants has been extremely useful to my project.

You have been so helpful concerning this phasing stuff. You got me started with a 1st generation autosome map not too long ago but things went down hill. So I decided to try the Xs. First thing I discovered using the gedmatch graphics like you showed me was a crossover in the 1st generation group that was not being picked up by FTDNA or gedmatch using the default settings. The matching area was less than 5cM! But it showed up clearly when I stacked the siblings up.

I think the hardest part for me when I tried the Autosomes was dealing with the fact siblings can and will match each other twice in many places. When I started I only had 4 1st generation siblings and the phasing maps looked pretty good to me. Then I had a close cousin to work with. I just keep going around in circles when I try to add the cousin into the map to determine which chromosome came from which side of the family. I must have missed something because things just won't work out right.

Now I have a number of descendants from additional 1st generation siblings (with more coming in thanks to the xmas sale and coupons - boy don't I love those coupons ;-). Since the new kits can only match the 1st generation siblings one way I may have a little better luck keeping things straight in my head. Or NOT! Only time will tell, but I am determined to try.

Do you have any additional words of wisdom?

I am moving this conversation over to the Family Finder Advanced topics since we are now talking about phasing the autosomes. This is a source of a lot of frustration. You are not alone. With practice, though, it can be a lot of fun to use siblings to figure out where you match each of your grandparents.