Amount of DNA vs Number of Matches

A Segment-ology TIDBIT

There are two concepts at work here – each is different – keep them straight and separate. Each in it’s own way can provide insights.

1. Amount of DNA from Ancestors. Although DNA from your Ancestors is random, there is some uniformity. You get exactly 1/2 of your atDNA from each parent; pretty close to 1/4 from each grandparent; about 1/8 from each great grandparent. By the time you get to your 128 5G grandparents, the average will be 1/128 from each one of them (roughly 1 percent), but with each generation past your parents, the deviation from the average increases. Still – the total of all ancestors at any generation will sum to 100 percent. On average you got about 15cM from each 5G grandparent, which may be in one segment or spread over several segments. The companies report that, in general, you will not share enough DNA to match most of your 6th cousins; but we know that you have so many 6th cousins that you will share with many of them – experience shows you will share from 0-21cM with some your 6th cousins. [see The Shared cM Project 2.0 at https://isogg.org/wiki/Autosomal_DNA_statistics ]

2.Number of Matches from Ancestors. This depends on several factors: Number of cousins, who tests, endogamy, etc.

A. The number of cousins you have from each Ancestor is directly related to the size of their families leading to the number of their living descendants. Larger families => more cousins => more Matches.

B. Similarly the number of cousins you have from each Ancestor is directly related to how far back that Ancestor is. Distant Ancestors => more cousins => more Matches.

C. Who tests. This may be most pronounced with recent immigration. The more recent an Ancestor immigrates, the fewer number of living descendants in the US, where the highest percentage of atDNA test takers live. Recent immigration => fewer cousins tested => fewer Matches.

D. Endogamy. With endogamy, the amount of shared DNA increases, resulting in many more Matches who otherwise might not have matched. The classic case is Ashkenazi Jews, who tend to get many times the number of Matches others get, and the Matches are in fact somewhat more distant than indicated. This also applies to other endogamous populations. Endogamy => more cousins => more Matches.

Some of my experience and observations

My maternal grandmother was from immigrants to US in 1850s. Roughly 1/4 of my atDNA is from her, and 1/4 of my Chromosome Map has TGs from her line. But, I get relatively few Matches on this 1/4 of my Ancestry, and some of the “gaps” in my Map have no Matches in them at all. When I get only 1 or 2 Matches in a TG on my maternal side, it’s pretty certain they are from my grandmother’s line. The other maternal TGs are from my mother’s paternal side with deep Colonial Virginia roots – they tend to have many Matches.

As the Matches pour in, and are added to my TGs and Chromosome Map, I can see where some TGs have many Matches and must come from distant Ancestors – perhaps at the limits of my Tree, or beyond – so I focus on the closer MRCAs (Most Recent Common Ancestors) which provide pointers. In some cases I’ve found the Ancestor had a Plantation and raised many children.

Some of the other TGs have small to medium numbers of Matches, and where I can find an MRCA, they tend to be smaller Colonial families – folks who moved around a lot; died earlier in life; and/or had few known children who survived.

So as you form TGs, and look at the resulting Chromosome Map, they start to paint a picture…

 

[22K] Segment-ology: Amount of DNA vs Number of Matches TIDBIT; by Jim Bartlett 20170517

16 thoughts on “Amount of DNA vs Number of Matches

  1. You raise many of the ideas I have pondered about atDNA matches. Whenever I see a high number of in-common matches I presume the CA is someone with ancestry in colonial America, When I see a match with a low amount of dna spread over several segments I wonder if we have multiple CA’s, probably tracing to colonial America. A match with a high-ish amount of dna and few shared matches, those I get excited about since it may lead to new information.

    Like

    • Barb, There are many possibilities, and the DNA is random. So it’s hard to form rock-hard rules. But you and I are in agreement that the number of Matches is an indicator that may help.

      Like

  2. I just started reading this blog and find the information very valuable, but am confused as to what the acronym TG stands for in this most recent post? And in the first comment is CA the same as MRCA?

    Like

    • Eileen, My apologies. TG means Triangulated Group. Usually I type the full phrase, and then use the TG abbreviation, but I slipped up this time. CA means Common Ancestor. Each TG has a Common Ancestor who first formed the DNA segment defined by the TG. Matches who are closer closer cousins on an Ancestor closer that the CA, will have an MRCA (Most Recent Common Ancestor). It is possible to have closer cousins on different MRCAs out to the CA of the TG segment.

      Like

  3. Thanks for that- you appear in a little TG with one of my McIntyre’s on Chromosome 21 I have about 4 Bartletts on FTDNA. Love to solve that one, we are still in Scotland.

    Like

    • Valerie – well you sure have your share of kits at GEDmatch – and I only match one. Our TG is about 15cM and includes 22 Matches (including my maternal uncle), but no other leads on the Common Ancestor. Send me an email off line about the BARTLETTs – I am the Admin for the BARTLETT Y-DNA Project, and have a lot of info about all BARTLETT lines.

      Like

  4. Scott, Mr. Bartlette’s son is a 3rd cousin of me through his deceased wife. He does this stuff for a living. He is very helpful and could probably answer questions or give advice. However, he is very technical!

    Have you made contact with your 1/2 cousin Swhartz? Could be your uncle or cousin.

    Love you, Pop

    On Wed, May 17, 2017 at 11:00 AM, segment-ology wrote:

    > jim4bartletts posted: “A Segment-ology TIDBIT There are two concepts at > work here – each is different – keep them straight and separate. Each in > it’s own way can provide insights. 1. Amount of DNA from Ancestors. > Although DNA from your Ancestors is random, there is some unifo” >

    Like

  5. JIm:
    I have a TG on chromosome #1 that goes from position 234.6 to 247.2 and includes a total of 28 kits. Four of these match my kit on a segment of 29.4 cMs. Many of the kts match from 234.6 to 238.0 while some of the kits match from 238.0 to 247.2 and other kits match somewhere in the middle of the segment. Obviously, the kits matching postion 234.6 to 238.0 do not match the kits matching from 238.0 to 247.2. However, all kits match my kit by definition, and all kits match the four kits that match my kit from 234.6 to 247.2. Is this one TG group or a number of TG’s?

    Like

    • Charles – sorry for the delayed response – it’s a complex question. The answer is both!! You have new crossovers and TGs at every generational level. At the parent level you get 22 large TGs – one for each chromosome – clearly there are intermediate crossovers and TGs within each one. In one respect this example is trivial. But this example at the parent generation is made so we can then think about the TGs at the grandparent level. You can read my blog post about crossovers, but, in general, let’s assume the average of 34 crossovers in each generation (a little less for males and more for females). This is an average, but you probably won’t have as low as 15 or as many as 50 in one generation. We’ll use 34 for talking purposes. So your paternal chromosomes (from your dad) will have 34 crossovers scattered over 22 chromosomes, resulting in about 56 segments. This is 56 TGs with first cousins – about 2-4 on the large chromosomes and usually one on the small chromosomes – sometimes 0. Of these 56 discrete segments from your grand parents, only about 34 of them will be subdivided by crossovers going back one more generation – so you’ll have about 56+34=90 TGs at the great grandparent level.
      When you are sorting out TGs from your shared segment data, most of the time the crossovers are fairly easy to detect. But what we don’t know is at what generation these crossovers were formed for the first time.
      So in your case: 234-238 is probably a TG and 238-247 is a TG; and 234-247 is a TG at a closer generation. So the smaller TGs may involve Common Ancestors way back – maybe in your Tree, maybe beyond it. The Matches who span both TGs must have gotten their shared segments from a closer CA.
      A chromosome map can be made for each generation – and each closer generation has some segments that combine the segments from the previous generation – see the charted in the Bottom-Down blog post.
      Hope this helps. Jim

      Like

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