A Segmentology TIDBIT
Ancestry’s ThruLines is like “dumpster diving”… sometimes you have to dig through the trash to find the pearls. Sometimes there is a smorgasbord of various genealogy junk, but sometimes there is a treasure trove of good information. Pick and choose wisely…
[22CF] Segment-ology: ThruLines Levity TIDBIT by Jim Bartlett 20240211
segment-ology 
I think “dumpster diving” is an excellent way to describe Thru Lines. One thing I’ve seen happen fairly often is showing a match that shares say about 800cM of DNA, and Ancestry predicts them to be something like a fifth cousin, twice removed. That pretty much screams “NPE” and is a good sign that I’ve found a treasure in the dumpster.
LikeLike
Glen, Like most things in genealogy, we have to use judgment. Usually 800cM of DNA is pretty close to a 1C, but I get the drift of what you are talking about. Often just a cursory look at the ThruLines line of descent will shout “look out”! When I dig into those – some are really way off track, but some just skipped a generation, or got mixed up on a given name like Susan Nancy Mary Thelma JONES, and just a little deeper research by me points to a much more logical Ancestor. I call these a “fixed TL”, and move on (ThruLines did help me). Jim
LikeLike
The first thing that I write in the notes is the cM count and segment count. ie 23-2 but I’m looking for the 1 segment matches. Using 1 segment shared matches you at least have a chance of looking at potential “triangulations” even though Ancestry doesn’t offer a chromosome browser. And sometimes you eventually find someone also at GEDmatch or MyHeritage or FTDNA.
LikeLike
Ray, I’ve tried several “first things” in the Notes over the years. I’ve finally settled on the Ahnentafel # & Surnames of the closest Common Ancestor. For me, the thing I want to do most is group the Matches into my family lines. I also have some special Dots: one for a Triangulated Segment (when the Match is also at another company); one for a Common Ancestor (ThruLines or otherwise); one for an incorrect ThruLines; one for a conflict between the Ancestry “side” and a TG or ThruLines (something isn’t in alignment); one for Tagged Matches linked to my Tree (I’m way behind on this one, but it’s important because when I enter a Matches line of Ancestors from them to our Common Ancestor it tends to bring in more ThruLines); one Dot for any line for which the Match may be, or know a close relative, carrying the Y-DNA or mtDNA of one of my Ancestors. Some of these I’ll want to revisit the list.
LikeLike
Thanks Jim.
<
div>This is part of the many disclaimers I will be claiming
LikeLiked by 1 person
What I don’t understand is that Thrulines has incorrectly inserted an NPE Great Grandfather into my Maternal Maternal, Maternal line—this despite no DNA proof. My proof is in the DNA where five testers on my Mat, Mat, Mat line: myself, my brother, my first cousin ,and two 2nd cousins, all match one another. And we all match another family (same place and timeframe) undoubtedly the connection to my unknown GGF. It’s baffling that Thrulines ignores my DNA matches, and my tree, and instead uses in Thrulines a GGF found in an undocumented tree, without DNA matches to my family. I’m starting to doubt many of my matches in Thrulines based on this error. And there is nowhere to go on Ancestry to rectify this error, plus it will undoubtedly circulate.
LikeLike
Linda – I understand and agree with you. ThruLines used (before 2023) to be pretty accurate – I’d say over 95% of mine were correct. In the past few months, however, their algorithm has gotten much more aggressive. By that I mean it will “match” with a wider range of dates, does not seem to take location into account, and various name spellings. And in the past month I’m seeing more and more that are based on just two Trees with no records in them – the “data” may actually be a typo, and they are passing it along. When I do an actual search I can find 200 Trees with almost exact same data (which matches my data) and yet the algorithm chose to use the flaky “evidence”. ThruLines is not a human, it’s an algorithm, so I don’t really know what’s “under the hood. But I’ve learned to click on the “evaluate” links and reject flimsy links.
If anyone knows how to rectify these errors, please post it. I do two things:
1. I refuse to enter suspect links in my Tree
2. I sometimes post a “comment” in the “evidence” Trees and ask the owner to correct or delete the issue so ThruLines doesn’t use it.
Jim
LikeLike
I have had wonderful success with ThruLines. I frequently write to DNA matches with unlinked trees, encouraging them to link their DNA results to their tree – not only to get their own ThruLines but to index their surnames making it easier for me to examine their tree.
LikeLiked by 1 person
Greg, I, too, am an overall fan of ThruLInes – they can easily report links that would take me a long time to find (and forever in a private Tree). I also write to DNA Matches with unlinked Trees when I can find the Common Ancestor (usually from a Shared Match imputation) – I add them and their Ancestors back to our Common Ancestors in my Tree and then ask them if they agree. This also has the effect of indexing their Ancestor names in my index. Trying to do this with all my Common Ancestor Matches – about 1,000 done and 7,000 to go… Jim
LikeLike
Thank you. The new “ancestors” on mine are mostly reliant on trees that have a lot of guesses and very little, if any, documentation. If ThruLines weighted DNA clusters over trees, it would be much more effective and might even find brick wall ancestors.
LikeLike
Good insight – I agree! I put my ThruLines info in the Notes; and then those Notes show up in Clusters. The Clusters are based on a kind of DNA “magnetism” that brings them together. The Clusters vastly outweigh any random, undocumented genealogy – and gives us a clear direction to follow. I agree with you and rely on Clusters to find bio-Ancestors. Jim
LikeLike
Jim, can you explain the magnetism? I’ve noticed when I start adding notes to a match with only 5 shared matches. The next day there are more. Does Ancestry deliberately hold back information?
LikeLike
Ray, I don’t think Ancestry ever holds back information. Sometimes a Match adds to or changes their Tree. Sometimes folks switch around who they link to their DNA, and then switch back to themselves – this also changes what you see. Ancestry is always trying to show you as many Matches and ThruLines as it can figure out.
DNA magetism: Not like a true magnet, more like an affinity (we are drawn to certain foods, smells, colors, etc.). We have a list of DNA Matches – each one shares some DNA with us. Match A will also have a Match list. Our Shared Matches are the ones we *both* share – each of these SMs could be from any part of our Tree. When 2 Matches have the same SMs with us they tend to be on the same area of our Tree. When 10 Matches all have most of their SMs on our Lists, about the only way that can happen is if we all have the same Common Ancestor. Usually we all share the same DNA segment, but that’s not a certainty. For me this is the “magnetism” of DNA – drawing our Matches together into groups – Clusters or Triangulated Groups. Jim
LikeLike