GENEALOGY-DNA-L ArchivesArchiver > GENEALOGY-DNA > 2012-03 > 1332382974
From: Jim Bartlett <>
Subject: Re: [DNA] Phasing 4th Cousins
Date: Wed, 21 Mar 2012 22:22:54 -0400
References: <4F6A49C2.email@example.com> <REME20120321183300@alum.mit.edu><C77D5C96-82B9-4382-B733-ECDA769836DF@verizon.net><REME20120321215901@alum.mit.edu>
Thanks - I'm thinking of a patchwork effort. In the end we have the whole quilt. And I think we would wind up with 2 quilts - one for maternal side and one for paternal side - and this would be true for each person in our ancestry. Another way to look at it is that any particular point on a Chr would be covered by a maternal segment and a paternal segment.
If we limit this to large segments, known to be IBD (say over 10cM each), the paternal/maternal quilts would have a lot of small holes.
For genealogists, this is OK - we would be focused on these large road signs (not concerned with the small gaps). By comparing with our atMatches we could eventually follow these road signs to virtual ancestors - just beyond our brick walls. What a vision...
Jim - Sent from my iPhone - FaceTime!
On Mar 21, 2012, at 10:01 PM, (John Chandler) wrote:
> Jim wrote:
>> If one were to try some phasing by large (IBD) segments that is
>> attribute these segments to specific ancestors - couldn't that be done
>> with somewhat fewer testees?
> Let me see if I understand what you're suggesting. Suppose you have
> two people with a known common ancestor pair and a large segment that
> is half-identical between them. You can hypothesize that the segment
> is thereby phased and IBD and attributable to one of the pair of common
> ancestors. Yes, that's possible, but if the shared ancestors are many
> generations back, you may very well not find any large half-identical
> segments. That's why you need an exponentially growing number of
> testees to get a reasonable shot at commonality.
> John Chandler