Family Clusters – Those Almost-Identified MRCA’s

This began while reviewing my son’s matches.

There is a website called DNAGedcom.com with an application called ADSA – Autosomal DNA Segment Analyzer.  It’s brilliant.  It does all the hard work of identifying matches which can be triangulated.  Running that application listed my matches chromosome by chromosome and showed me exactly where two or three or even more were matching on the precise same segment.  It could be the same segment on either side, but that can then be deduced.

When I ran my own, I found it interesting that I have lots of matches on some segments and not many on others.  For example, on chromosome one I have a segment of 8-10 cM which is matched by twelve people.  Two of those twelve match for exactly 8.76 and when I look at their profiles it’s obvious that they are siblings managed by the same person.  They match each other for much more but that 8.76 cM is what I have that matches them.

Running ADSA showed many how many of my matches belong to such groups and are obviously from the same family which has tested several people.  I collected them together as much as possible and realised that I have less than half the distinct matches than I thought I had, because being siblings the match is basically a duplicate for me.  There’s no point individually deducing my MRCA with each of those chromosome 1 matches, for instance.  I just need to deduce their own MRCA which is their father, and call them one match.

I felt much better.  I didn’t really have 380 matches, I actually have about 105, and my son has even less.

Tree

So first I grouped them together and called the family groups A, B, C,D etc.  Then I worked out which segments I share with each family group.  On chromosome 11 on the same segment I match fourteen cousins with Prince Edward Island ancestry, most of them in the Normanites Group.  This is not coincidental since group members are trying to test as many relatives as possible in the hopes of identifying their certain connection.

When I looked at my son’s data using ADSA, I discovered an absolutely huge block of same-segment matches on chromosome 17 of 8 – 14 cM in length.   FORTY of them!  Many came with family trees and looking through the trees, as usual, I failed to find a common link.  So I looked at the profiles and found a few managed by one person.  I emailed her and asked if her profiles belonged to a DNA Project since my son’s kit was matching them so well.

They didn’t.  However, she ran her own ADSA report and picked the probable MRCA  for her own matches in a very short time.  The majority of those forty profiles are in the United States and have links to the surname Garner.

John Garner and Susanna Keene were a couple in early Virginia.  Susanna was born in Virginia but John emigrated apparently from Shropshire in England.  Final confirmation of his birthplace has not been made.  John was born around 1633 so this is a long long way back.  It would take a particularly persistent segment to come down to the present day, but a few cousin matches might help it along.  In my son’s line, this was also helped by a few elderly fathers so there were not as many generations as you might expect.

It was an enlightening couple of emails for me.  One thing which has stayed with me is that many have not uploaded their whole family tree to FtDNA.  No doubt this was because one could not edit an uploaded gedcom, one had to delete and then upload a new one.   I also learned about ‘cold spots’ being segments from the distant past.   John Garner was born in 1633.  Either he himself was the MRCA or it was even earlier.

I looked into it further.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s