A Segment-ology TIDBIT
Are Clusters based on Common Ancestors (CAs) or Triangulated Groups (TGs)? I said CAs, and Jonathan Brecher said TGs. I now think Jonathan has the best answer. My apologies for doubting his conclusions.
My point was that with a few, large, close, Clusters, each Cluster must be formed on a CA, and include many TGs. A Match Clustering which results in 4 or 8 or 16 Clusters (which they tend to do) are clearly formed on 4 grandparent, 8 Great grandparents or 16 2xGreat grandparents – and this is born out with the Leeds Method and other experience. These large Clusters must each include many TGs – and my experience bears this out.
However… There’s often an “however.” As the Clustering thresholds are decreased, the number of Clusters formed are increased. In my recent example I had 156 Clusters using a 20cM threshold. Note that one fourth of my Ancestry is from 1850’s immigrants, with very few Matches (and all of them were close cousins in one Cluster). I should have had about 208 Clusters. And 161 Matches did not cluster. This gets us pretty close to 256 6xGreat Grandparents 8 generations back. And a number of my 156 Clusters appear to link with only one TG [Note this is AncestryDNA data, most of it without TGs].
I am now reviewing my AncestryDNA Matches and trying to assign Cluster IDs to each one, by looking at the info I have in the Notes box for each Match and reviewing all their Shared Matches (and their Notes – easily viewed with MEDBetterDNA). In most cases, where the Match shares a single segment with me, I’m tending to identify a single Cluster. And when I have TG information, it’s tending to be one TG.
So, I’m going to eat some crow, and apologize to Jonathan Brecher. I now think he was on the right track, and that we should try to link Clusters to TGs (specific DNA segments). After all, each TG is from a specific Ancestral line. Of course, at AncestryDNA (without segment data), we’d still Cluster mainly on CAs. However, with a comparison between AncestryDNA Clusters and Clusters with other companies (with segments and TGs), we should be able to find a correlation between our AncestryDNA CA Clusters and TGs. Through this correlation, we could “impute” TGs to AncestryDNA Clusters.
So – thank you, Jonathan Brecher – for Clustering for several of us, for your comprehensive analysis and for your insight!
If anyone has been Clustering around CAs, that is still OK. Think of your Cluster CAs as potentially having multiple TGs – particularly the closer CAs (4C-6C range). And as you run Clustering with smaller thresholds, and find more Clusters, you’ll find your former Clusters subdividing into smaller Clusters, which smaller Clusters would tend to match up with one TG – Walking the Clusters Back!
[22AE] Segment-ology: Clusters Link to TGs and an Apology TIDBIT by Jim Bartlett 20190222
I have a cluster of 5 members from an analysis of FTDNA results (done using Genetic Affairs). 4 out of the 5 share a chr 8 segment with me that I share with an Ancestry match (Bob). 3 of them are on Gedmatch so I can compare them with Bob and we form a TG. Bob, who like me is Australian, is connected through my paternal grandfather’s branch. We share a 2G-GF whose parents were Irish. (We don’t share the 2G-GM as there were 2 marriages). 3 of the matches in the TG are Irish, and one even has the same surname as Bob and my Common ancestor (although it is a common name). So everything is pointing towards this cluster being linked to this TG from an ancestor of my 2G-GF.
However the 5th member of the cluster, Sue, does not share that segment with me, and is connected through a different branch – my paternal grandmother’s branch. Marie is also Australian, and the ancestor I share with her is also Irish. Bob has no known connection to my paternal grandmother’s branch.
How would you interpret this?
See below. I’d “go with the flow”, and look around for a reason why Sue is in that cluster. Maybe she does share a different line with the other 4 – it’s technically possible.
I have one cluster with 5 members from an analysis of Mum’s FTDNA results. 4 of the 5 triangulate on a Chr 8 segment that Mum shares with an Ancestry match (Bob). Bob, Mum and 3 of the FTDNA matches are on Gedmatch so I can compare them all there and they form a TG. Bob and Mum are connected through Mum’s paternal grandfather’s branch.
The 5th member of the cluster (Sue) doesn’t match Mum on the Chr 8 segment. Sue is connected to Mum’s paternal grandmother’s branch. I can’t figure out which branch this cluster represents.
With Cluster’s we often see things which don’t make sense. Clusters are based on ICW Matches – which ICW Matches could come from a wide range of Common Ancestor. If you have four Cluster, each one “tends” to come through a different grandparent. But a problem with Clustering, is that Matches are placed in only one Cluster – but some of our Matches share multiple segments with us and there is always the possibility that the multiple segments will be from multiple Common Ancestors (including both sides of our Ancestry – a have a few dozen such Matches). So when an issue comes up, I look back at all the TGs each Match is involved with. In fact, I mentally try some Matches in different Clusters, to see if there is a better fit. However, if most of the Matches all have the same TG, I’d got with that. Jim
Jim, just another of my nuisance questions. Is this correct: Clusters and Circles are NOT scientific, but TGs ARE scientific? Thanks, again.
Vance – AncestryDNA Circles are a special form of ICW – ones that include matching Ancestors in Trees. Circles are not necessarily a TG.
Vance – Thanks for the kind feedback.
My opinion is that these segments need to be Triangulated like others. Yes they could be from a crossover event – either in your DNA or their DNA. You both got a segment of DNA from a Common Ancestor – and for a match to occur, these segments overlap. Usually one or the other starts first, and one or the other ends first – it’s rare for cousins to have exactly the same segment. What you “see” in a chromosome browser is only the part that overlaps. So the two ends of an overlapping segment are caused by a crossover point in your DNA and/or your Match’s DNA. I try not to predict a probability of Triangulation. If I get a new shared segment, and it easily “fits” inside a paternal TG, but would span across two maternal TGs, I would guess that it was a paternal TG. But I always check!! Sometimes my guess is wrong. If, like at FTDNA, you cannot compare the Matches against each other, look at their ICW lists – if they have several ICW Matches which are in one TG, then the probability is very high that this Match/segment is too. Jim
LikeLiked by 1 person
And one more question if I may – Ancestry has their function called “Circles” – with maybe 6 or 8 people and lines connecting any particular person to others (but maybe not all) and choosing another person can change who is connected to that person. Is this creating any TGs or just illustrating “in common with” data?
Excellent series, Jim. I will need to read several times to get the full effect.
For now, a question. What is your opinion on the significance of several matches starting at the same address on the same chromosome and falling under a longer segment starting at an earlier address? Can these segments with same start address be from the same “event” (maybe a crossover)? Is there a high probability that they will triangulate?
Jim, am I correct that TGs are “proven” with a one-to-one; and clusters are just “proven” with a tree match? Thank you!
Yes, a TG will follow only one ancestral line. You could have Matches with MRCAs all along that line – as different cousinships. For Clusters, it depends on the threshold and number of Clusters you got. If you got 4 Clusters, each one could have multiple TGs, and each one would be on an Ancestral line through the same grandparent (for that Cluster). Clusters are formed from ICW or Shared Match lists – a Cluster is a group of Matches who match most of the others (in ICW lists). There is no knowledge of Trees or DNA segments involved. It just works out that a bunch of Matches who all (mostly) match each other, do so because they have a CA on the same DNA segment. That’s why they are on each other’s list. In my case I have 380 TGs covering my DNA. I expect that with 380 Clusters there would be a one-to-one for each. With fewer Clusters (which most of us get), there will be some Clusters with multiple TGs. The Clusters will generally, roughly align with CA at a generational level. If you get 64 Clusters, plus or minus some, I’d expect each of them to align, approximately, with one of your 64 4xG grandparents. The DNA is random, so it usually doesn’t fall into exactly equal groups, so don’t try to “force fit” Clusters. You’re looking for a general trend.
Jim Bartlett – atDNA blog: http://www.segmentology.org
Very interesting, Jim. Do you think if two or more TG’s have the same CA, that they would be in the same cluster?
Interesting question. Would? Or Could? On one hand I just said that at the limit, one TG per Cluster. And some closer (fewer number) Clusters could have multiple TGs. But you can have TGs on different chromosomes from the same CA. You will have many TGs from a grandparent Ancestor – all over your DNA. However, if you are talking about the tips of the quills on the genetic “porcupine” fan chart – each quill would be a distant Ancestor, with a specific TG, and, with a low enough threshold, have one Cluster. However, the closer, thicker, parts of a quill represent multiple distant Ancestors, TGs and Clusters. So it’s hard to form a firm rule – too many variables.
Jim Bartlett – atDNA blog: http://www.segmentology.org