GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2012-07 > 1342214563


From: BARTON LEWIS <>
Subject: Re: [DNA] IBS vs. IBD - Family Finder Results
Date: Fri, 13 Jul 2012 17:22:43 -0400 (EDT)


Thanks Jim. When using FTDNA's Chromosome Browser, I guess one should
be looking for segments of shared DNA between persons one believes to be
descended from a common ancestor. When the segments are small, this
gets kind of dicey. In one instance, I noticed 2 different matches
shared the precise amount of very small segments of DNA on 3 different
chromosomes, and thought that was potentially revealing.

Barton


On Fri, Jul 13, 2012 at 11:02 AM, Jim Bartlett wrote:

> Barton,
>
> Yes. It will probably take me about a year to get most of my genome
> mapped, and you can be sure that I'll report back what I find. I'm
> very interested in the 7-10cM segments:
> 1. I think that they are roughly the limiting size for most atDNA
> algorithms, and real cousins with Common Ancestors further back will
> have only part of this segment, and therefore not pass the algorithm
> and thus not show up as matches. So the most detail (the most distant
> ancestors we are going to find with the current atDNA matching
> algorithms) are in this range; and theoretically we should be able to
> map our whole genome with them. And that's the end, using the current
> processes.
> 2. I'm interested in seeing if on one side or the other of many such
> segments is a segment from a close cousin (the other side being
> completely random)
> I believe that all the sutdies will confirm that there is very wide
> variance, and you can make universal rules as you are trying to do -
> think about my football field analogy.
>
> For your second paragraph question, it's too difficult for me to tell.
> I think each segment has it's own history. To understand this, you
> have to think about the track the atDNA takes. You start out with an
> ancestor, say 7 generations back. This ancestor recombines large
> segments of atDNA from the 2 Chr01 he/she has into one Chr01 that
> he/she passes to a child. Repeat for all 22 Chr. What gets passed is a
> full set of Chr01-22 to the child, which has roughly half of the
> ancestor's DAD and MOM in it. The child gets the other set of Chr01-22
> from the other parent. Suppose in this process of combining large
> segments from two parents into one Chr, there were some small gaps.
> The process tries to insure each Chr is 100%, so it might gather up
> some loose ACTGs and fill in. I'm not sure of how all this works and
> am just guessing here (I'm an engineer, not a biologist). Anyway, this
> same process happens again for the next child, who also gets a 100%
> set of Chr01-22 from the ancestor !
> we started with, but it's a different mixture of the ancestor's two
> sets of Chr01-22. So each of these two children has roughly half the
> same segments as the other one, and roughly half that are different.
> As you continue down the generations of descendants of each of these
> siblings, you can see that the segments from the original ancestor get
> smaller and smaller; and you can see that the amount shared between
> cousins at each succeeding generation gets less. It's actually quite
> amazing to me that after 6, 7, 8 or so generations there is anything
> recognizable as a matching segment left. So would two segments from
> different lines of descent make it all the way? Statistically, it
> could happen. On the other hand we probalby got a number of segments
> from our Great-grandparents over a number of Chromosomes - and some of
> these would be shared by 2nd cousins. So it's all a matter of degree
> and probabilities. I know this: don't make any blanket guarantees
> about any of this.
> Paragraph 3... I would not say "almost certainly". I would contact
> every match as if they were a cousin. You are going to find many for
> whom you can't find a Common Ancestor due to the fact that the two of
> you don't know enough of your ancestry to make the connection. I think
> this percentage is going to far outweigh, and in fact mask, those that
> are IBD, and we may never know the answer to your question. I wouldn't
> dwell on it - dwell instead on contracting your matches.
> Jim Bartlett
>
> On 07/13/12, BARTON LEWIS<> wrote:
>
> Jim, thanks for your follow up. Wouldn't it be great if there was a
> study of matches whose exact relationship was known to extrapolate
> from their shared DNA the probable relationships between matches whose
> relationhsip is not known?
>
> Maybe you can answer this question: when two people have matching
> segments of DNA on many different chromosomes that are greater than
> the smallest segments reported by the Family Finder test, are those
> the result of the relationship to the MRCA? Or is it too difficult to
> tell at that level?
>
> If two people have 2 large segments (5+ cM or greater) of shared DNA,
> are those almost certainly the result of the relationship to the MRCA?
>
> Thanks, Barton
>
>
> On Thu, Jul 12, 2012 at 1:12 PM, Jim Bartlett wrote:
>
>> Barton,
>>
>> My best advice to all is to ignore the predictions, don't try to
>> guess the generation, etc. You just get bogged down into trying to
>> make something precise which has a very high range of possibilities.
>> It's like trying to find the high spot on a football field... (OK,
>> over the top, but I hope you get the idea)
>>
>> Summary: Communicate with your cousins, share ancestries, agree on
>> Common Ancestors - it's that simple.
>> More advice - don't try to guess which atMatches will turn out to be
>> good cousins - communicate with them all! You never know how they
>> descend from your Common Ancestor.
>>
>>> Jim Bartlett
>
> -------------------------------
> To unsubscribe from the list, please send an email to
> with the word 'unsubscribe' without
> the quotes in the subject and the body of the message


This thread: