Can someone help me with this?

[IgorStrav]

Please forgive me for this query, which certainly demonstrates my (still) somewhat sketchy grasp of DNA and Ancestry DNA matches, and is possibly something which I should be able to work out for myself.

However, your help appreciated.

All of this refers to Ancestry DNA testing.

I have a 4th cousin DNA match with M which is 'extremely high' at 62cM. 
I can see from M's tree some matching ancestors and therefore think I know from which branch of my tree we are connected.  Her Gx3 grandparents are the same as my Gx3 grandparents.

A new shared match with M is H, at 66cM. 
H has no family tree, but in addition to the shared match with M, H has shared matches with other people also descended from the same line.

So I am relatively confident that even without a tree, I know from which line H and I are connected.

I was recently trying to find a link with person P, with whom I share 57cM. 
P does not have much of a family tree, and the paternal line is also missing, because of illegitimacy issues.

I therefore did a bit of research for P on the maternal line. 
Whilst I found a common surname for me with P, I could not link the person with this surname to its appearance in my own tree.  It is likely to join at some point, but much further back than a 4th cousin link suggests.

With the new appearance of H on my DNA matches, I now find that P and H are appearing as shared matches with me, but P is not a shared match with M, or any of the others I see from this line.

I am now torn between two courses of action

 - to approach H (who is a very recent member of Ancestry, has no tree, and may have done the DNA test for non-family-history-research reasons) to see if I can get any further information on what might be her connection with P or indeed me.

- to continue to research the known line with M to see if anything pops out.  I know the full names of both H and P, and would therefore recognise them if they appear.

But why might M and P not be 'shared matches', if H and P are and H and M are, and all of them are with me?

And I apologise for the length of this, and if you have - thanks for reading

[diplodicus]

My two children both share 3,450 cM with me. My first cousins each share 800+ cM with me. Their grandchildren (first cousins twice removed) share about 100 cM. That leaves a lot of cMs that have disappeared randomly in reproduction.

I would assume your connections with these three individuals are from a common ancestral source. It is quite feasible to have three significant matches where two of them them don't share a common strand with the third one.

I have used Excel to map these matches. I have one group - my Ingram branch - that has sixty matches in it. The two strongest matches are not shared with each other!

Two things to bear in mind:

• Ancestry states that autosomal testing only finds about two-thirds of matches.
• Not all legal relationships are also biological ones!
• Strong matches don't always show a connection to weaker ones; especially when the weaker is <20cM

[IgorStrav]

Thank you, Diplodocus.

I was coming to the conclusion that a spreadsheet was required.  Have just been pondering the best way to set it out.

Family history is endlessly educational, isn't it.

 

[Rufous Treecreeper]

I'm learning the DNA stuff too 

I had a strong 4th cousin match but it turns out, thanks to the match having a public tree attached  that we are 6th cousins who just happen to share a predicted 4th cousin amount of DNA.

So it's very possible your link is much further back.

Hope this helps,
Marlene

[IgorStrav]

I'm learning the DNA stuff too 

I had a strong 4th cousin match but it turns out, thanks to the match having a public tree attached  that we are 6th cousins who just happen to share a predicted 4th cousin amount of DNA.

So it's very possible your link is much further back.

Hope this helps,
Marlene

Thank you!   

lovely.........

Endlessly educational.

[diplodicus]

Things I have learnt along the way:

I use the "notes" function to record shared matches, any conversations, and the date when last reviewed. If the match has a tree I also record how many people it contains (so I can easily see if the have added anybody).

in my spreadsheet I give each 'bunch' of matches an identifying letter (e.g. I = Ingram or JW = Jonathan Walters). This is essential as you will soon find more than one bunch.

You need to be the first name in both the first row and the first column. This is vital so you can then use excel's filters to highlight any particular cluster (you filter on "your" column since this is the only one that contains all your shared matches).

Try not to be tempted by 5th to 8th cousins until you have worked your way through those sharing > 20cM (i.e. 4th to 6th cousins in ancestry-speak).

Highlighting the starred match icon can be a very useful marker but try not to overuse it otherwise it becomes difficult to find that match to which you really meant to return. Once you have looked at a match, then little blue "new" icon disappears but not until you leave the match. If you wan't to keep it as "new" you will need to click on it twice (i.e. turn it "off" and then back "on").

Being someone with far too much time on my hands, I also record the cM count alongside their name. This helps me find the correct row in which to insert a newly-arrived match (I have 269 matches > 20 cM).

Every time you open your spreadsheet, make a backup. It is depressingly easy to mess it all up. Also, become an addicted presser of the "save button".

When Ancestry adds a new strong match, it doesn't alert you to new "shares" with your existing but weaker matches. Sometimes, I have to trawl back through them all to see if anything has changed. This is where my notes become invaluable.

At first I only recorded those 4th to 6th cousins with shared matches but slowly more of them have become a cluster, especially once I entered the <20 cM realm. I suggest recording all strong matches.

[Gadget]

Hi Igor

I'm not sure if you've used this before but it might help show you the range of possible relationships with a range of cMs

Just put the cM reading in the box and the probable relationships will show:

https://dnapainter.com/tools/sharedcmv4


Gadget

[Pheno]

Just marking this so I can come back and look at the spreadsheet in more detail - looks good.

Pheno

[IgorStrav]

Just marking this so I can come back and look at the spreadsheet in more detail - looks good.

Pheno

Yes, it does, Pheno, doesn't it.  Thank you, Marlene (Diplodocus), that's a great help.

I shall start doing this.

In good news, I've finally had a response from my DNA match P giving me her Ancestry User Name as well as her email address,  which means I can invite her to see the research tree I did.

Given I have her full name and also H's full name from Ancestry, we can see whether the names ring any bells in the present day.