The Debate Continues Over Usefulness of Small cM DNA Matches
HTML-код
- Опубликовано: 17 июл 2024
- A viewer asks whether the multiple segments of shared DN below 7 cMs are useful in finding relationships. Well, he's generally wrong. Let me show you why you're getting false matches.
⚖️ More evidence that small cMs aren't valid 👉🏼 • Are Small DNA cM Segme...
----------------
CONTINUE LEARNING
📺 Are small cM matches FALSE MATCHES? • Are small cM matches a...
Show Notes: isogg.org/wiki/Identical_by_d...
↪️ Grab your FREE genealogy research guides and templates at www.familyhistoryfanatics.com...
⚡ Want to grow your family tree faster? Join our membership program.
/ @familyhistoryfanatics
📗 Like to read? Check out these books
www.familyhistoryfanatics.com...
----------------
CHAPTERS
00:00 Introduction
00:48 Are Small DNA Segments False?
04:46 Are Multiple Small Segments False?
06:31 Devon and Andy aren't related, or are they?
09:44 More evidence against small DNA Segments
12:46 Final analysis of small segments
----------------
✅ Let's connect:
✔️ Subscribe for more genealogy tips: tinyurl.com/FHFanaticsRUclips
✔️ Website: www.familyhistoryfanatics.com
✔️ Share Video Ideas: www.familyhistoryfanatics.com...
#FamilyHistoryFanatics #geneticgenealogy #DNAResults 
Хобби
One of our aims in genealogy is to push our ancestors back in time, and DNA matching can help us with that. But as we go further and further back the matching segments tend to get smaller and smaller, and about the time that easy records start to run out is when the DNA starts to dip below 7 cM.
The conundrum is that we want to find matches who connect with us via common ancestors that far back, yet those matches may have genuine matching cM's below 7 cM with none at 7cM or above, and we would like to be able to detect them
In practice I've found that small DNA segments with the lower threshold at 3 cM can be useful for checking out people on our paper trails whom we feel we could be matching, but who don't match us at the higher 7 cM threshold. Typically I find that they do in fact share quite a lot of small segments with me.
One person at Ancestry didn't match me or my known matches at Ancestry or at Gedmatch at 7cm, but at 3cM she matched one of us on 53 segments adding up to 254 total cM! There are people who don't match me at 7 cM but match a relative of mine at 7cM, and match me on multiple small segments under 7cM.
Someone on my paper trail might not match me at 7 cM and have only one or two matching segments at 3 cM, but those who do match me at 7 cM tend to explode into multiple matching segments with me at 3 cM. There's a reason for that and that reason is that the matches at 3 cM are real matches, not false positives.
Overall - I know that really small segments can be false positives, but if there's a lot of segments shared with people on a paper trail , and there are quite a few people on the same paper trail matching each other at that level, then that's helpful to know.
You make a great case. However, in this video, you'll see how you can just as easily be misled by the sharing of DNA. So, it's something to be very mindful of. I recommend people determine how everyone above the 7 cM threshold is related before attempting to work with anyone below that marker.
I just carried out your exercise by obtaining kit numbers from people in my first cousin tree whose tree I did a few years ago. I relate to her on our mother's sides. I compared some relations on her father's side to me in Gedmatch's 3D chromosome browser. They definitely don't relate to me on any paper trail (like you my own tree is well developed going back a very long way and so is my first cousins's tree because I did it) yet the problem you outline has surfaced exactly as you describe. I don't match these people definitely unrelated to me at 7 cM yet at 3 cM I match them on heaps of segments. Therefore I must bow to your superior wisdom and give up on small segments!
If you have a documented relationship with someone and the cm shared is small, don't despair, test multiple family members, especially older relatives, who may share as expected. Build trees of your DNA matches on that family line, getting a more complete picture with both a paper trail and DNA. Some of my full sibling's matches that are over 100cm do not share anything with me, or just small segments. I don't know if it is true but sometimes my small match overlaps with a portion of my siblings' larger match, I take it as a good sign. People who aggressively dismiss smaller segments (even as much as 20cm) for others bothers me, we are all independant researchers, if you include them, fine, if you do not, that is okay too, for you. As your great examples show, the smaller the segment, the less likely it is from a documented time frame, as long as we are aware of that, we have done our due diligence
Even overlapping small segments can be false.
@@FamilyHistoryFanatics Yes, true
Fantastic video, as always! I love seeing evidence on why there is a threshold in which matches below that level don't count. I would imagine it is a different threshold for each individual looking at their DNA.
There is probably some slight variation between individuals, but not enough to matter.
This is one of your most helpful videos for those of us working to find cousins that will help identify third or fourth great grandparents as an MRCA. Your examples in the video clearly demonstrate the false match problem with small segments. I suspect that many researchers who are using genetic genealogy tools will face this issue eventually. Thanks again for another excellent video.
So glad you learned what you needed.
African Americans searching for their continental ethnic ties used the lower cm matches because the paper trails were non-existent due to the slave trade.
While specific familial ties can’t be held with strong confidence below 6cm, a person can pick up on trends in matches with certain ethnic groups.
If the sample sizes are fairly balanced, and you’re matching heavily with a particular ethnic group, it is quite likely that you are related to that group.
As more people take these tests, it adds more clarity to distant relationships.
Ethnic group calculations and matching algorithms are two completely separate things. I have not seen any research that would support a "likely that you are related to that group" other than the fact that you are from Africa and all African groups share DNA with each other (just like all European groups share DNA with each other).
Yes, the person uses match calculation. No, the person that is tracing does not necessarily rely on ethnic ‘calculations’.
We will see more research in this area to be sure, but that doesn’t mean people can’t embark on their own investigation.
Generally, people rely on the testimony of the match that they contact. AAs reach out to their matches to garner more information. It’s one thing to make broad assumptions based on a single testimony, but multiple sources of testimony will often lead to clarity. This is why I do not agree with Ancestry’s decision to cull matches below the 7cm mark.
Genealogy research is also about pulling clues beyond just genetic analysis. I hope you agree.
Surnames for example are a huge help in this context because West African surnames tend to be very localized.
That said, I can’t stress enough that larger trends are key when using this method.
@@FamilyHistoryFanatics You don't know what you're talking about when it comes to African Americans. YOU may not have academic research, but many of us have oral family history. My great-grandmother said our ancestors came from Nigeria. When I took my autosomal DNA tests, my largest African match was with Nigeria. When I looked at my DNA relatives section I saw that I had low cM matches with four Nigerians who were *ALL* from the Igbo ethnic group of Nigeria. A lot of Igbo people were enslaved in South Carolina. My family is from South Carolina. I think that from a preponderance of the evidence it's more than likely that we have Igbo ancestry. Please don't speak on us because it's offensive and because we have more information about ourselves than you have. Just stop.
Thanks for this very clearly presented and very interesting video! I have been really interested in small matches since I "discovered" a bonafide 3rd cousin from a "missing branch" of my paternal tree (my grandfather was orphaned at an early age and his line was lost to my family... until my finding.) I matched this person with only one segment of 7 cM (not "chopped" by 'timber' and before Ancestry's purge) and they had my surname in their tree linked to a recent generation. I researched the paper trail and it was undeniable (and absolutely fascinating find for me!). I faced alot of doubt from my family about it - I educated myself with the DNA painter and other resources, so I knew it was possible. Sure enough, my Aunt tested very recently and she matched the same user at 38 cM. This match, along with several others now, has now helped uncover my grandfather's whole family in New York city in the 1800's... (though sadly no records indicate my patrilineal 2GGF"s hometown in Germany, where every third person shares the Muller name...) Needless to say, I "saved" all of my 6 and 7 cM matches before the purge, primarily for this reason. I understand that many of them may be false matches; I don't waste much time on these; they've got to get me back to those specific places with some solid (and familiar) names/dates. HOWEVER if I can parse out some common information, a sort of independent verification of a link, I have some faith that it could be a "real" match. Even if the actual segment wasn't passed down that spercific line, if I can find a paper trail (and fortunately, there actually are some good records for early Dutch and English in certain places for the later 1600's), I feel there is benefit in being able to show a link on paper.
To sum up, small matches could be useful but they take a lot of work to validate. Correct?
Absolutely - they *might* take alot of work. But it is also possible, as it was in my 3rd cousins case, that the person has a name and/or a place that is familiar, and some simple digging can produce a paper trail connection. A 3rd cousin is fairly close - my surname was still on her tree in the past couple generations; I just needed a few documents in the paper trail and it was a sure thing. DNA painter shows that third cousins can range from 0 to 234 cM match. That means any 6,7,8,9 etc cM match could be as close as a third cousin. On the other hand, I have some 3rd cousins that have tested and we don't match. We don't doubt our trees, behind mountains of research and other matches, but it is possible that you don't share that DNA (at least not above the threshold of the site being used). I use the same approach to evaluate a 7 cM match as I do a 70 cM match. Statistically, the 7 cM could take alot more work, or it could be that outlier that is a ~3rd cousin. Ironically, I have linked some 7th cousins in the 30-40 cM range (also supported by DNA painter statistical distributions) Though, I have found most of my low matches related to my one Dutch line that goes back to the 1600's in New York state or English line to New Haven colony. There are alot of family history books and fairly good church records that can be used to reconstruct those lines (many of them already reflected by prior research); these are often 7,8,9, and 10th cousins! (I know I cannot guarantee that the shared segment is definitively from that ancestor, but on paper I can find a link; that without the indicated match I would not have known about. In other words, since the match and I shared a trace of DNA, I tried to find a link and could. There are many I can't - and I know where my threshold is to abandon that effort! :-D )
On Ancestry I have 44k Matches, but only 3145 have a shared match. I check for shared matches as new matches are added, and Identify the non-matches with the grouping function. With my 6-7cm matches I managed to keep Out of 22k only 600 have shared matches, but 30 of these have been useful as they have Common Matches, and have helped me identify family lines for higher matches with no or minimal trees. So I agree the result is small, but still useful.
I agree, Kay.
Before Ancestry reduced its cM threshold, I flagged everyone below 10cM who had shown up when I'd searched a fairly remote family name .
(Lines had been drawn, which excluded my NS lot from The Family. )
I needed to confirm that I was indeed descended from a certain Sk.. through his only child/son by his first wife (a Whi..)
Ancestry in ThruLines showed me a half 6th cousin , via the second wife - right on! (I'd not have pursued it, since it's a private tree.)
This happens to be one of my "strong " lines, so I have lots of NS matches, as well as an appropriate number- at appropriate distance- from their original area, NH & VT.
I've too many mysteries- - this isn't one of them!
This is another good technique for identifying matches who don't share other matches (and are likely candidates for being false matches). I should do a video on it.
Interesting debate I have & still look at as I have four autosomal tests & just taken a fifth after long discussions at RootsTech London. Now, I compare 1:1 Gedmatch kit numbers at 3cM & find at times very different responses & I also look at all the info at the bottom of said comparison i.e. cM, total % SNPS, Total Half-Match segments (HIR), shared segments ,. I know of the false match issue but each of my kits when compared can vary wildly based upon the test company & can have an 8.7cM largest match with one kit & 70% SNPs & against another persons kit 8.7cM & 53% SNPs. Also some companies will say no match yet another will say 9cM. I started creating spreadsheets to log data to facilitate analyse & look for trends.
Once you get down to looking at small segments, you can see lots of things that don't follow logically. That is why most genetic genealogist educators try to steer people away from them.
My brother has many distant matches that I don’t have. So the more people in your family you test, the better your chances are of finding ancestors further back. I found one line in my distant matches. But I found another line from my brothers distant matches. And both can be confirmed with records going back to the late 1700s.
Statistically, you only share about half of your matches with your brother and the majority of these are biased to the lower cM.
Does anyone know the Power book in the background? I can't read the rest, thank you!
Thanks for asking. The book is called Power Scrapbooking. You can learn more about it on our website before making a purchase. www.familyhistoryfanatics.com/books
Ancestry cut off is 8, and I think 23andme is similar. Then there are 3rd cousins that could share none. I have no choice but to deal with random pieces.
As a general rule, any match I have below 15cM needs to be anchored with other matches that are more than 15 cM before I will pursue it. Otherwise, you have an increased likelihood of barking up the wrong tree.
@@FamilyHistoryFanatics I do a lot of barking, so you're on to something.
I look at names and haplogroup and where their family is from. Most of them I have no idea how they are related.
I do notice the closer the match the less shared matches I see. Which I guess makes sense.
Small segments can be useful in doing dna mapping of ancestors. For example, say you have 3 siblings. 2 siblings match someone with a small segment of data (even if it’s a false match, if 2 siblings have the same segment, they will get the same false match). 1 sibling doesn’t. So with the two siblings, you know that they have the same ancestor association on their chromosomes. While the 1 sibling isn’t matching them. So these types of tidbits can in areas where you have little firm data, transitions points. Thus helping with segmenting dna maps.
Not really, because you have no idea how that false small segment was formed.
My parents share several reasonable cM matches on MH. And my dad shared DNA with cousins of my mother who don’t share DNA with my mother but paper trail back confirms the link. Probably different lines but because of these multiple signs of connection between my parents when I come across a dna match on my kit that they do not share I check if I share multiple matches they do share dna with and the chance that maybe they just share below the cut off for matches but I share enough from them that it pushes me above it. For example: 7.8cM for them each and 12cM for me. We also have matches that don’t make sense as a recent connection so unless we locate a documented trail back to connect and multiple common matches that tally the results I put those matches in the false match pile.
This is a special case, but to even see this you would have to have both of your parents DNA (which you do), in which case, there really isn't a genealogical reason to be looking at your matches (except to look for this special case).
@@FamilyHistoryFanatics yep that’s exactly what I do :). I use my dna to identify those outside the box to see if it ties up those inside the box if I see several cross over signs. I usually stop at 20cM for specific research blocks but grab all of them to identify potential links between the groups further back in time.
"Our last common ancestor is 30k years ago"
That seems unlikely, if that African-American friend has European and likely English ancestry, the common ancestor must be far more recent, plus from what I've heard mathematical models say that the last common ancestors of ALL humans is around 2000-1000 BCE.
Edit: You addressed his European ancestry, but I wonder your opinion on those mathematical models. We likely didn't receive much if any ancestry from this recent common ancestor but the mathematical models seem solid given that they make very conservative estimates.
That's right! From mathematical models the first common ancestor of all humans lived around 2000 years ago. This model is explained in more detail in the video "every baby is a royal baby" by numberphile.
@@dmfaria83 Yep, even if you were to exclude the most isolated populations in the world the date would more recent than people think.
You're correct, it is closer to 2000-3000 years ago (I have a video about that as well). In this case, I like the hyperbole to emphasize that we are not at all related in a genealogical timeframe.
You are exactly spot on here.
Lots of small segments, especially as low as 3, are not an indication of relatedness, in any genealogical timeframe.
One of the problems of the constant debate about this is, I think, a matter of scale. And many have a 'story' about the one match that was relevant. [NEVER as low as 3 that I have seen.] "Yes, statistics, But I know they can be *useful* stuff...."
I've got one, at 6.8 that was real. Solid paper trail to 5th ggrands. Triangulated solidly with larger matches of 20-40. No endogamy or recent-ish pedigree collapse. A bingo.
But an accumulation of tiny bits? Naw.
Unless you are one of the super-pros working intensively with, say, Acadian DNA, the tinys are a waste of time.
And I'm not sure that they are really on a productive path....
I don't know of anyone working with endogamy matches that would touch small segments. Mid-size segment (7-15) are already a problem for them as it is.
I have a similar situation where I share no segments >7 cM with all but 1 of my 5th cousins on one line or to my uncle with those same cousins. Not too surprising. However, Using a second cousin male on that line to me there is a perfect yDNA match to a male 5th cousin on that line (37 and small variation to bigY500). And although the paper trail is suggestive of this match through two brothers to a common father, the yDNA gives us the proof that we need. This is a reasonable process, right to verify.
I sadly match my husband at 7 something cM and show an MRCA of 7.2 I think or so on Gedmatch :( My parents I also discovered, were I think 4th cousins as they each descedned from a brother in the same family. However, Gedmatch says my parents were likely not related recently and DNAgenics says my parents were not related. So
Many people are more related to each other than they realize.
I find every match I can add to my tree useful, because eventually I will find another match that is important and sometimes the common ancestor is shared with the wee little match I added months before.
If you can add the match to your tree, then yes, it's useful. Just be aware that some matches are false and can't be added to your family tree.
What about my segments that are longer than 2 others and My Heritage chromosome browser doesn't say it triangulates but it is above 7cm? (I rephrased the question from chat, because what I had originally didn't make sense) Do they have to start and stop in exactly the same place? I have some that start in the same place but the end is longer?
I use MyHeritage and was wondering this as well.
Often, I’ll look at a match and we share segments with many people on the same area of a given chromosome but most of them aren’t displayed as triangulated matches because the shared segments don’t start AND end in exactly the same location. It didn’t occur to me that it might be related to the length of the segment. I wonder if this is a case where I should apply the common sense rule and assume I’m related to all of these matches or is there more to it than that?
No they don't need to start and stop in the same place. If they aren't triangulating, then the overlapping amount may be less than 6 cM or they may not match each other (i.e. they are related to you through separate parents).
@@FamilyHistoryFanatics thank you!
Hi . I am curious about your comment that you share the same haplo group with your brother in law. I am wondering, did you just test 111 markers or did you do the big Y 700? There is a big difference in how far they break down your exact haplo group when they test 700 markers over there not testing 111 markers. But I expect you know that.
Actually, taking a larger test wouldn't have helped to resolve the issue. Have you seen my video where I specially talk about my brother in law and are shared haplogroup? ruclips.net/video/OcgDYF_VcOA/видео.html
@@FamilyHistoryFanatics No, I have not seen that one. Thanks for the link. I will give it a look.
I understand small segments are not reliable and I also understand that multiple small segments totaling 40/50/60 cMs are likely noise. I would be curious on your opinion on multiple small matches 30 plus for instance with a total cMs of over 120-150cMs is this still not a significant amount??
Yes. It may be something that happens in endogamous populations (I don't know because I haven't worked with endogamous populations), but that would probably be outside of a genealogical timeframe.
"...anything that matches the child but does
not match one of the parents has to be a false segment..." Unless it's a false negative for one of the parents.
That requires further evidence to substantiate. Generally speaking, if the child is a biological child of a set of parents, the matches that do not match the parents is a false match.
I tend to agree with you in regards to the low segments, but I don't discount the match itself. Sometimes the lowest matches have great trees and are useful for that, if nothing else.
I do have a question. Even if the match survives the phasing with both parents, there is still a chance that it is a false match, even a false match for the parents, isn't there?
Yes, but it is much smaller chance. Probably the most common small segment false match is when it is creating the matching segment by taking parts of matching DNA from the mother and father (some of which match each other by chance as well).
@@FamilyHistoryFanatics Ok, thank you for clarifying this for me. I also meant to say earlier, thanks for another great video.
Andy - I had a MyHeritage match of 3 segments with the largest being 29cM. Tester matched me but neither of my parents. MyHeritage said "recombination". So how do I know whether ANY match is a real match? Maybe a big match to one of my parents is a similar bad match. I know I've spent a lot of time building out other people's trees and not finding any matches there. So if a large segment is suspect, how would a small segment be good.
The better question to ask yourself is how matches with 29cM segment don't match your parents. I would be willing to wager that this is the only one. And you may have a few dozen others that do match your parents. So lets say 2% of your time looking at large segments is wasted. I can live with that.
On the other hand, how many matches with the largest segment being 6cM don't match your parents, and you will find, more than half of them. In other words, 50%-99% of your time looking at small segments is wasted.
@@FamilyHistoryFanatics it wasn't a 29cM match it was 29cM in ONE segment plus 2 other segments. That's a pretty big segment match to be false. This happens a lot on myheritage. They tell me it is because my kit came from ancestry and my parents' kits are from ftdna. I don't know the origin of the "matching" kit. I've had several problems with triangulation where A=B and B=C but AC on the same segment with the same start and end points. They've examined and acknowledged the false matches but blame it on the kits being from different companies.
I'm starting to believe that Ancestry has the advantage in that they aren't comparing apples and oranges by not allowing outside kits to be uploaded.
I can't understand it... Is there anybody that could take a look at my DNA test and help me to understand it? Please!!!
What are you trying to understand? The ethnicity results or who is related to you?
Lots of tiny common segments doesn't mean they're not related since we don't match all our cousins
It means that the DNA can't be used as evidence of a relationship, you are stuck with paper records.
So how about just one segment match? What is the threshold there?
How big does a single DNA segment match must be for there to be a likely family connection?
Me and a friend only share one DNA segment but of the length 43cM (which seems significant) but with many cousins 3. degree I have multiple segment matches but each segment is much lower than 43cM.
What is the meaning of just having one large one?
Not much. The vast majority of your matches only share a single segment with you. I am not sure, but this sounds like a good thing to research, I may have to put together a survey to get some data and see what we can find out.
@@FamilyHistoryFanatics Let me rephrase my question to make it more concrete.
Imagine you have two matches:
Match A: 4 matches with 10cM segments
Match B: 1 match with 40cM segments
Both share 40cM in total.
Are both similarly closely related to me? Or is one of the two more likely closer related?
Intuitively, I would say that there seems something more special about 1 match with a relatively large segments, because I would expect the breaking up into smaller segements.
In my matches, this one match with large segment seems rare.
Also, maybe inbreeding in a population could play a role? or i would not be surprised that some segments are stickier and don't break up as easily due to physical or biological constraints.
I have no deep expertise in quantitative genetics and never looked at it deeper. I am just guessing here but I do have a PhD in theoretical physics.
I do agree that a lot of small segments does not prove a match. That said, I feel that small segments in and of themselves should not be overlooked, if you are able to view them through a phased kit. At that point, I think that the small segment may have more significance and I will pair that with checking for shared matches between the 2 kits. I have a 3X Great Grandfather from Ireland that I am trying to break through a brick wall to find out who his parents were. This has been a 20 year quest and I still don't know. The descendants here in the United States from that 3XGreat Grandfather do not have information and so I have tried to find descendants of the 4X Great Grandparents in Ireland who might be able to help. Unfortunately, the DNA evidence at this level is tricky because most of it is small segments with only a small amount having larger segments. I have been fortunate to be able to identify some of the Chromosomes that are relevant to this 3rd Great Grandfather and in combination with using the phased maternal kit, along with shared matches, although not 100%, I think I have been pretty good with dismissing most of the false positives of those small segments and retaining the ones that show they are most likely true matching small segments. Unfortunately my brick wall persists because of the lack of information in what trees are available, the lack of response from "matches" and the lack of records for the area my 3rd Great Grandfather appears to be from during the time period for which he was born.
I did a video recently. ruclips.net/video/gs8DwFXnFUw/видео.html
The method you describe makes the small segment similar to 7-8 cM segments as far as false matches. An improvement, but not a big improvement.
I finally got one of the DNA test by a well know company, I am 70 now. I will mention I know quite a lot about my family history. I mentioned the percentage in a RUclips comment section and got jumped on for that. The person saying how grossly inaccurate they were. What say you, did I waste my money? Because the percentage mentioned did back up what historical research had let me know.
Percentage... do you mean ethnicity percentages or shared DNA matches?
To be truthful, I don't recommend people may much attention to the ethnicity percentages. The databases are too small in many countries to provide accurate results. Even then, these are estimates.
If it aligns with your family tree, awesome. If it doesn't, maybe there is a reason.
I wouldn't jump on you for believing the percentages. I would caution against tatooing the percentages on your arm.
@@FamilyHistoryFanatics I used Ancestry and it gave me a 56% Scottish number. Which based on what I know about my family is about close. I would have thought between 48% and 52% just from family history.
But a question arises, if sites like the one I used are unreliable have we been sold snake oil?
Small segments can be valid, but personally I would assume they aren't unless they triangulate. In the video, when you compare yourself and your wife and find several
I did that (just didn't show it in the video), none of them were the same or overlapped.
Small cM matches may not confirm relatives by themselves, but can't they help confirm that people in your tree actually ARE you relatives? I have an ancestor in my tree who I could not confirm 100% until I started getting distant DNA matches for some of her other descendants. This ancestor was 5 generations back (3X ggrandmother) and I first got a 6 cM match, then found a few with a bit more cM, and finally I found a match at 24 cM. There was no mingling of these families after that generation.
You would have to show that the small segments weren't by chance (which almost no one does). So no, it can't confirm that people in your tree are your relatives.
Very helpful information! Thanks so much because this topic is entirely relative to my brick wall. Is identifying a 4g-grandfather through 5th cousins unlikely, and have you ever known or heard of anyone who successfully employed this technique alone to discover their 4g-granfather?
Using small segments to solve genealogical problem? Several people have claimed to, but I have yet to see a case where the small segments were the key evidence in solving such a case.
@@FamilyHistoryFanatics Thank you so much for responding, sir. It's an incredible long shot, but I'll never know unless I try. Last quick question: should I go with a Y-DNA test and join a surname project, or try autosomal DNA? Your advice is much appreciated.
@@jasonborn867 If you are in the male line, yes! Y dna is the telescope that can focus on that distant star. If you are not in the paternal line, find and pay for the Y test for someone who is. For everything else, autosomal is better.
I am working on a 4th grandmother like this. I talked with that pretty blonde geneticist last year. She said to have a better idea I needed to find as many in the gen above me (ppl in their late 70s) who have taken a test. Not much luck there. No one ever emails you back. However, I have gotten leads from names in people's Ancestral names list. My problem with Ancestry upping cms to 7 is that it doesn't take into account those of us who have extensive accurate trees for generations with locations and names those locations. That's when smaller cm segment helps. Esp when you take out pile-up areas. Is the solution a full genetic genome(?) I wonder.
@@sl5311 It's tough finding solid leads when going back so many generations, but I agree future technology should factor in the variables you mentioned. I'm still trying to decide if y-dna or autosomal testing is best, and not sure either will provide results given the time frame in question. Hoping the technology will get better!
I solved a mystery using Y DNA….AND 2 distance autosomal match about 9 cMs each, that turned out to be sisters of our Y DNA matches’s 3x great grandfather. I also had that same thing happen when I Y tested a maternal cousin. I found 10 distant matches, about 6 cMs each, all related to siblings of my 3x great grandfather! So I swear by distant matches.
This is an example of the survivor bias. You are remembering and highlighting the ones that were a success without also explaining how many others that were distant never turned out to be solved with a small match.
@@FamilyHistoryFanatics No, I understand that. I am just saying that it is worth looking at small cM matches if there is something else that leads you to them. I searched for people with a certain surname or location in their trees and then began to work each and every match and see who else they matched with, if any one at all. I also knew I was looking for relatives of my 3x great grandfather so I knew the match would be small if it showed up at all. Plus I knew those matches most probably only shared one 3x great grand parent with us.
I've asked this once, and you've already answered it, but I have to ask it again.
How possible is that I have many matches with whom I share the 30cM biggest segment, but neither my mother nor my father is related to any of them. In addition, the children of these persons are already relatives of either my mother or father.
Ilona Egyed Lászlóné (Mészáros)
Kor: 70 - 80 között
Balázs Egyed
által kezelt DNS
Kapcsolatba lépés vele: Balázs
Becsült rokoni kapcsolatok
3-5. szintű unokatestvér?
DNS-egyezés minősége?
0,7% (45,8 cM)Közös DNS
3Közös szegmens
30,1 cM
Ilona is not a match for my father, but Balázs' son is.
Fun fact: Balázs is a genetic professor and the main employee of Hungary's market-leading genetic laboratory, but I haven't been able to agree with him about our kinship yet because he doesn't answer LOL
It is possible (not very likely) to have a false segment that is that large.
I have come to the conclusion that small CM's are not extremely helpful. With two exceptions. One: With me when the match is also on my grandparent's list. Because who my grandparents are related to, so am I. Two, when there is already a paper trail, and this may be a way to determine relationships between 7th and 8th cousins, or even further back with a paper trail.
Yeah, that may be helpful. But you would need to make sure that you have your segments mapped through a Leeds chart or painted out to further ancestors.
If we all share almost 100% of our DNA, how is it possible to tell a part each person through DNA? Can you explain what are shared CM?
I have covered parts of this on several videos, but it sounds like this would make another good video so I will put it on the list.
Disappointingly convincing.
Sorry?
@@FamilyHistoryFanatics I have used small cMs to try to break brick walls. You have convinced me that is probably a wasted effort which disappoints me but I do thank you for teaching me that I am not doing a useful thing.
I'm looking for specific matches on specific chromosomes as a way of noting possible members of an ancestral family. The higher the SNPs, the better. I have some examples of why I do the research that way. Even though the success rate isn't high, it's not like I don't have enough spare time to spend on it. So let each person research the way they want and don't encourage the "it's just noise" bro science.
I appreciate you participating in this discussion.
My goals stem from the hundreds of thousands of comments I respond to every day. Too often people focus on working in the noise and get frustrated when it doesn't pan out. Instead, I hope people will focus on figuring out all their matches above 7cM are relate before going down the rabbit hole that is fraught with more false matches than not.
However, not everyone will follow my counsel and that is their decision. At least they will know why I discourage it.
I'm CONFUSED! HOW is it that you can share NO DNA with a KNOWN third cousin but we are to disregard all cm less than 7 as not valid? Make it make sense
You never answer questions so what’s the point?
Didn't know that I needed to be more clear.
The point is, that below the 7 cM threshold, you're chances of figuring out how small segments are related to you increase. And, just because you have many small segments in common, that doesn't necessarily mean you're related. Spend time on the larger shared cM matches first before going down this rabbit hole. Does that help?