Am I on the right thinking here.

[SouthseaSteel]

I guess endogamy that exists within certain communities and that includes Ireland e.g. west Mayo and multiple relationships e.g. parents being cousins or suchlike are different in nature but manifest themselves similarly.  In any case, both %^&*er up conventional cM analysis

[Essnell]

Hi Everyone,  Well i got the weighted shared DNA info  thankyou david. 
thus:  for person's half brother:   Shared DNA: 976 cM across 37 segs.
                                                   Unweighted Shared DNA: 976 cM
                                                   Longest Segment  77 cM
                                        Pedicted Relationship:  First Cousin  or Half Uncle.

     For Persons Full Niece, Sister's Daughter:   Shared DNA:  1867 cM across 42 segments
                                                                      Unweighted Shared DNA :  1867 cM
                                                                    Longest Ssegment:  119 cM.   

since last here I have separated the match list for the niece into parent 1 and parent 2 according to Ancestry. 

this was interesting: Many more for Parent 2 than Parent 1.      Parent 2 people I have already tracked and they predominantly lead to the grandmother's family.  Where the missing father is concerned most do not have a tree online. 
Parent 1 group are looking more like the group belonging to "person wanting to know"  their mother. So I have two female lines befuddled by the looks of things.

That is as far as I have managed and laptop is in need of charging.   be back in a while. 

  Essnell

[Essnell]

Hi  up and running again.....

Is it possible that this is being caused because there is nothing on any thing about either the father or the grandfather.  Not anything on any tree that I have seen.

I am not a DNA expert but my common sense tells me that there has to be DNA in someone's kit test that is being ignored.   
It has to be in the DNA for the 'person wanting to know ' .  It should also be in the Niece's DNA  ,  one would think. 
 That same DNA will not be in the half brother's test.   I wonder how Ancestry is dealing with this.

I also know that there are three half siblings to the grandfather provided I have got my searching correct. 

Cheers Essnell.  off to do more thinking......

[David Nicoll]

Esnell,

    Hi, I think you are thinking along the right lines.

    Am I correct in understanding you have the following

    A, B - Sibling
    C - Half Sibling
    D - Child of B.

    What are the relative ages of A,B,C? Is C the eldest, if so the DNA may be correct?
    I would be looking at the shared DNA between C and D, what is that predicted relationship?
    Ignore the known relationship and go with the DNA for the moment, confirming relationships between shared matches of A, B, C and D individually.
    I have not used it myself, but many others have, the WATO tool on the DNA Painter web site may help here.

    Happy Hunting

[Essnell]

hi David Nicoll,
Thank you for your reply.    yes A is the one wanting to know. 
                                                B is his full sister... same parents both sides
                                                C is half sibling to both A & B    same mother different father.

                                              D is dau of B  so niece of A. 
   ages approx as i really do not know the exact dates.   A  early 60-65
                                                                                       B  mid 60-65
                                                                                       C late 60 - close or in 70.
                                                                                        B 25-30
The relationship between C & D  is stated as:   Half 1st Cousin 1 x Removed or 1st Cousin 2 x Removed.

  I mentioned this fact earlier :

  A & B ' s father Q who was born in 1922 has 5 half siblings,  all born between 1935 and 1950 . Their mother did not marry until 1935 , 14 years later. 

I am not able to find any information about these people except names.  All family trees are private.   
 
So now to get these shared matches and examine those. 

Cheers Essnell.

[David Nicoll]

The light may be dawning!

So to clarify,

A to C 976 cM
A to D 1867 cM
C to D around 200 cM

It’s a bit early in the morning, but I think that would be what you would get if A and C are in fact cousins, not half siblings.
Does this make any sense?
If C has matches to A’s fathers side then that will make life a bit more complicated.
As you say shared matches.

Happy Hunting

[Essnell]

Hi David Nicoll,    I have just got a partial comparison of those Shared DNA.

A  to C   14%  976 cM     37 segs
A  to D   27% 1867cM    42 segs

C to D   309  cM   16 segs Longest 52 cM

B has not tested  she may need to. 
C to A   Nothing on the match page re A
C to D  similar to A. 

So C appears to have no DNA to either A or D.

Now I have just read your post as well.    : According to what I have been told emphatically is that A's mother had a child before she married A & B's father. That child was adopted out by her.  Her family have always known but A was never told. He only found this out last year. 
We have been looking for A's father for about 5 years with no success.     This half brother visited our place last Christmas. That is all I know.   But if this is the true DNA story that screws everything for a lot of people. 

That does not change anything about looking for his father.  Might be another search looking for who C's father was and his mother. 

So C looks likely to be not a close relative.
Been thinking  : drew a map   What if the adopted out child married and had a son ....    that being C  which makes him a Half First Cousin. ... which is what is coming up on Ancestry.  !!!!!

so one problem looking solved but another on the horizon but for much later.     

Still hunting for A's grandfather.     
Cheers Essnell.

[Biggles50]

OK, you have made great progress so far.

I would build separate family trees for the shared matches who appear to be the most useful i.e. have user names that can ID them and useful linked family trees, working back to each GG GP if you can.

I’d work on one match at a time, ignoring the others until I had between 4 & 6 family trees.

I would then look are the surname makeup of theses trees and using Pro Tools look at the cM shared between the people in the trees.

Where the trees have a link I would join them together in Roots Magic to form a new combined tree and export the tree back into Ancestry.

Make a note of the cM value of each of the DNA matches in the tree and then export said tree as a Gedcom file.  Import the Gedcom into DNA Painters WATO tool and enter the cM value of each of the DNA matches against them in the WATO tool, complete the presented question and let it start the analysis.

You will be presented with a list of weighted probabilities of the answer to the question.

Yes, endogamy can and no doubt will skew the results but it has to be worth a try.

As often written DNA can be a waiting game and Summer is a very quiet time for finding new DNA matches for me the last two weeks have only gained a handful of new matches, come the New Year they will no doubt ramp up again.  So do not get disheartened if no resolution is found simply wait a while and look again.

Given the cross fertilisation that seems to have happened a possible scenario is that there will be multiple families from where the unknown parent is.  If you can then identify a person who is a descendant of that unknown then approaching them and offering a DNA test could reveal what you seek or it can help in pointing towards another family where this process can continue until there is a resolution.

This is how I found my Half Sister and ID’d my Biological Father.