Y DNA usefulness?

Question

Hey, guys so I've been considering getting both the Y DNA and the mtdna tests. I know one of the biggest advantages is finding your haplo group and the migration path they took. I just realized that 23 & I does that too though. Is that just not accurate? The other advantage is to confirm your ancestor with someone else way back. In the cousin section it looks like I have a few cousins that have both tests. Was this groundbreaking to your genealogy research? On a side note what does the 400 dollar test do that the 100 dollar test can't?

Comments

25 April often has sales, as it is DNA Day.

Answer 1

Sale started at (link is so Wikitree gets credit for referral) Family Tree DNA: http://www.tkqlhce.com/click-3382244-13710348

Answer 2

I advise you to read over the DNA testing help pages, but short answer:

• Y-DNA - check your paternal lineage. I used this to disconnect my ancestors from what was listed here and other places on the internet, because it didn't line up with other testers. Very useful for deep lineage following surname (mostly, unless there was an NPE - Non-Parental Event)
• auDNA - verify your paper trail with cousins, by checking if you match at the expected levels. Useful for confirming triangles of lineage between two testers. (helps to determine where the paper trail might be questionable). Can also lead to other ancestors you are stuck on, that someone else may know in their tree when you match with them. I just matched pretty closely with someone last week with an entirely new surname, and appears to be on my maternal side where I have less testers, so things will pop up over time with new information even if it's not there originally. 
• mtDNA - maternal line - this one is sort of just entertainment value. The mutations in the mtDNA are so few and far between that any "matches" here mean basically that we shared a common ancestor anywhere in the last 1000 years for each mutation. On an answer on stackexchange, Harry V explains it as "a perfect match could be your mother or your sixth cousin, a distance of 2 could be your 5th cousin or your 40th cousin"  (https://genealogy.stackexchange.com/questions/11739/how-far-back-is-a-mtdna-match-of-genetic-distance-2-likely-to-be ).  We can bucket vast swaths of the human population into these maternal haplogroups, but that's about it.

As for the expense of the tests, the more expensive ones typically test more locations, so can tell you more specifically about how you match. Each of the companies use slightly different ranges for the testing, and different assumptions to fill in the gaps. If you want to match to existing testers, it might be best to use the same company. BUT, you'll find different matches at the other companies too. If you search for posts by Edison Williams, he explains it in-depth, and suggests the best idea is Whole Genome Testing, but that's expensive ($1k-ish US). 

Try to wait for sales (Mothers Day/Fathers Day are typically good times, other major holidays), and check to see if there is a study out there already that might sponsor some of your cost. There are certain branches of families that are looking for testers and will foot the whole bill if it's for an interesting branch, or fund some portion if you go through them.

Lastly, you may find that taking the test didn't necessarily tell you anything new, but just told you that you are probably correct about what you know. Valuable information, but something to be aware of.  

*edit - didn't realize my brain listed my explanation of auDNA under an mtDNA heading, so fixed that and added one for mtDNA before anyone notices...

Comments

Y DNA testing is not much like au testing at all. For one thing, it is a team sport, and somewhat expensive. True, we sometimes buy au tests for older relatives. But in the case of Y, without other testers, your own results won't mean much, which is why FtDNA wisely created the ability to gather together testers and compare. So first, please join some groups! Second, as you do your genealogy research, as the advice of the people running your groups about how to choose the best testers, and which tests you should buy. Personally I want everyone to do Big Y, but the 37 will show whether or not your tester should upgrade or not. Spend your time and money where it will answer your research question. Usually that's au -- and remember that you can upload your Ancestry.com au results to FtDNA, *and* a gedcom. Use that family tree and FtDNA will work for you.

Valorie

Answer 3

I did a y-DNA test with ftDNA over a year ago and recently did the Nebula Genomics 30x Whole Genome Sequence test and transferred the results to yFull.

My experiences:

1. I have not found any particularly useful existing yDNA matches on ftDNA. Their database is nowhere near as large as somewhere like Ancestry.

2. I did get one Pavey 5th cousin to test on ftDNA. We were not in the same haplogroup at all which indicates an NPE somewhere. That is useful information but a bit disheartening since I have traced the Pavey line back 14 generations and may not be biologically descended from them.

3. I did the Y111 test on ftDNA but that just gives me a Haplogroup of E-M35 which is a fairly broad (non-specific) group. They want you to spend more money of special tests to give you a more accurate sub group. It you want the most accurate haplogroup it works out cheaper to do Nebula Genomics I have been told.

4. Most of my yDNA matches on ftDNA have Morrocan or North African heritage since my Haplogroup is a subclade of E-M81 (the "Berber" haplogroup). This is an interesting mystery but yDNA goes back so far it could be a) I'm descended from a Roman soldier with ancestors from North Africa or b) my distant paternal ancestors were Moors in Spain and then moved to the Netherlands before moving to England (some hints of this in ftDNA matches). My ethnicity estimate on Ancestry only includes England and North-west Europe, Scotland and Norway. So my paternal North African roots are a long way back (1000's of years potentially).

5. It can be hard to interpret the results on ftDNA. The combination of Markers tested and Genetic Distance is confusing to me:

111 Markers: I have no matches

67 Markers: I have 7 matches and the closest genetic distance is 5 steps

25 markers: I have 253 matches and 3 of them are a genetic distance of "Exact Match". I have no idea how far back our common ancestor would be.

Comments

To Rob's point, the number of markers tested gives you more accuracy in the results. If you have 10 matches at 25 markers, that's meaningless if you have only 1 match at 111 markers. You're not going to find anything common in the other 9 people from the 25 marker level, the higher level of testing is saying that you aren't really that close to them.

Example: I have two good matches at a 111 marker level, with a genetic distance of 5 (meaning 5 markers out of the 111 are different). They created a new SNP for the three of us because we were closer than anybody in the SNP that existed prior to us taking the test. But, that doesn't mean we found out that we are second cousins or something that close.

The three of us have our family trees built out and are 95% confident in the results back to 1700 (eight generations), without being able to figure out who our common ancestors are. We think based on the number of mutations that we probably share a common ancestor sometime between 1500 and 1700, but we have some work to do to figure that out (if we even find it to be possible). We are all far too far apart for autosomal testing to show anything overlapping.

So it's interesting, and changed my focus on which line within the family to research, but doesn't give you super recent, immediate information.

---

I have zero Y-DNA matches at any level, (12 - 25 - 37 - 67 - 111 markers). My earliest known male-line ancestor was from present day Germany. 

---

While Lincoln may be from Mars, apparently, his experience illustrates that it depends on other people who match you having taken the test. You might be the first, and just sit out there all lonely for years before other people match up to you.

---

You shouldn't give much weight to an "exact match" when it's based on very few markers. A genetic distance of 5 at the 67 marker level could easily be from a common ancestor 250-300 years ago so may not be that useful.

---

I went into the process 'a little blind' and as a result was disappointed with what the results were. I'm adopted and looking for BF.  So with 19000 new results from Ancestry, was expecting a similar sort of breakthrough. I got very little (and even less) useful material.. However, I now understand the process a lot better, and part of me still says it was a waste of money. However, I realise that the more people that test, the better the results will be in the long term. After all, imagine being in the first 1000 people to test autosomal with Ancestry. "You have no matches and your ethnicity estimate is 100% human"

---

@Rob: My Y haplogroup is E-M183, also part of the same "Berber" haplogroup family. It's a complete mystery to me as well, because I have a hard brick wall on my family progenitor, with only whiffs of hints that it might be German ancestors. But with the history of that haplogroup family it makes zero sense so far.

Answer 4

Aside from the haplo group and distant anthropological historical knowledge,

• Generally, due to the miniscule Y-DNA test taker population, it is highly unlikely that you will find any meaningfully close living relations.
• If your male line is of a culture which used patronymic surnames in the very recent past (e.g. Swedish), then it may be completely useless. Your 30 or so Y-DNA matches will likely all have different surnames, leaving you with no gain.

That said, I had an adopted great-grandfather of unknown parentage going into it. My Y-DNA test identified my biological surname, which in turn helped me identify my great-grandfather's mystery parentage.

Answer 5

23&me gives you a Y-DNA haplogroup - a very broad grouping, of limited use for matching.

FamilyTreeDNA (ftDNA) and others do STR testing, which is more useful for genetic genealogy.  Additionally, ftDNA does "Big Y".

https://isogg.org/wiki/Portal:Y-chromosome_DNA

Even tho Y-DNA STR testing is useful for genetic genealogy, it is still limited - cousins that it identifies may still be generations away.  It can be very useful for identifying a surname for those searching for paternal-only lineage.

I've had good results with Y-DNA testing for my father's line (back into Ireland) and my mother's side (proving a PGM line).

And, as with autosomal testing, the database size makes a big difference.  ftDNA is a large database for Y-DNA, but it is not the only company that does testing.  I believe Y-SEQ may provide another large database?

Answer 6

(Incomplete answer, since others already covered most of it)

But expanding on two points  and clarifying a misconception that is one of my pet peeves.

1. Different haplogroups for one test are like different versions of the same address. 23andme might tell you that a tester Bob lives somewhere in Utah. A specialized test at FTDNA is telling you that information, plus his town and street address. Sometimes, you need that extra info.

2. There are two different types of mutations on the Y chromosome measured by FTDNA. The cheaper tests just measure STRs. The BigY measures even more of those, but also is the only standard test they offer that measures SNPs. Either STRs or SNPs can be useful in their own ways, so for some people, testing both is worth the money.

However, there is a cheaper option to get SNP knowledge. You look at other people’s SNP tests, and if you are lucky, you may be able to guess some SNPs that you would test positive for. There are multiple companies who can test just a few SNPs for you, and it is much cheaper. My father had success doing this and identified that he is positive for a mutation that is the “terminal SNP” of a couple of other men who took the Big-Y. This is the deepest knowledge publicly available about those men.

But if my Dad takes BigY, and I do plan to spring for it someday, he will have other non-public mutations that could help distinguish him from those men further. You can’t get that data from targeted SNP testing like he already did.

3. mtDNA tests are not just for recreational purposes. I have sprung for these tests for relatives and will continue to do so. Here’s a write up of one of my biggest mtDNA success story, which is more success than I’ve had yet with even Big-Y testing. 

https://www.wikitree.com/wiki/Space:Journey_with_mitochondrial_DNA

Any test can either prove useful, or not useful, and you don’t know until you do it. Even if you don’t find immediate use, if you wait long enough things might change.

Comments

Barry, you combed through 127 people, got one tester that matched, and you even have a discussion (below) about how there could be thousands or millions of people matching. That you happened to find one with paper trails to match is exciting, but is illustrative of how it is mostly recreational, because it takes an enormous effort to follow it through. 

There could be thousands of people in the world today who would show as a match to my mother if they tested, maybe even millions.

With such a large quantity of people, I suppose one could argue, "your test matched this woman in Missouri, but that doesn't say much because there are tons of people in this world who would match you — you can't rule out that she was just a match by chance, not because your theory about the lines that connect you was correct." But this logic is faulty. It is clear that the vast majority of people who test their mtDNA are not matches to my mother, even distant ones separated by 1, 2, or 3 mutations. The chances of a random person I contacted in Missouri being an exact match by coincidence are miniscule. 

I think the logic above (I italicized it for easy identification) about it being a chance match is indeed something that cannot be ruled out, given the amount of pedigree collapse that we continue to see in all families. It's at least something that must be considered as a possibility, although the preponderance of evidence in your case seems to support your conclusions (which is all we can ever really hope for). 

For a new tester who is just trying to figure out what to test? mtDNA is not likely to be very helpful.

---

It was actually much more work than you suggest, since I the process involved no "combing." I didn't look through an existing list, but rather I had to research lines down from my proposed ancestor, document them well and add their profiles to WikiTree, until eventually I had built that many profiles and found some living people to contact). So yes, it is a lot of work, and not recommended for someone just starting out in genealogy ... but that by itself does not constitute an argument that mtDNA is solely for recreational use. For some people it won't be useful other than telling them a deep haplogroup, but there are lots and lots of people who have gotten more than that recreational use out of it.

My objection to the dismissal of mtDNA that I see frequently and broadly is (1) there are lots of people who are serious genealogists who are only just getting into DNA, and they should not be put off from any potential useful tool, and (2) part of the difficulty with using mtDNA as compared to Y-DNA is numbers; the number of mtDNA testers is far fewer than Y-DNA testers. I applaud the Million Mito Project for working to increase the profile of mtDNA testing and hopefully increase the numbers of testers. But they are competing against a much louder group of people shouting down mtDNA, many of whom go much farther than "recreational" to call it "worthless", "a waste," etc.

So whenever I see these claims, I feel the need to speak up.

(p.s. IMO pedigree collapse in families is irrelevant to research with non-recombinant DNA . The structure of the tree of all womankind is a pure binary tree with an ancestor as node and living matrilineal descendants as the leaves. Pedigree collapse is a byproduct of examining trees in reverse, with a living person as the node and distant ancestors as the leaves, and the observation that if you go back far enough, the structure is no longer a binary tree.)

---

ok, I concede the point, as I was not attempting to call it "worthless" in any way, just difficult.

As for the pedigree collapse idea, I guess I was saying that it's *possible* that the true inheritance could have come from some other pathway than the paper trail shows, and yet still match because they were a matrilineal descendant from another path. But yes, terminology was incorrectly used there.

---

Great points Barry. I especially appreciate your distinction of the mtDNA (and yDNA) tree as binary. Some groups show strong endogamy within the genealogical time period. That's a problem of too many matches, many of which are not "closely" related at all within the last 500 years.

Answer 7

Much of what you're goig to read concerning BigY is likely now out of date information. Why? The Big Tree has doubled in size in just the last couple of years. 23andme will give you a very generalized haplogroup...useful for tracing migrations paths..as you say.
BigY700 is now comparing over one million SNPs. We've seen in the FTDNA Campbell Project several SNPs datig as recently as 1800 BCE at the apex of the bell curve. This is well within the genealogical timeframe and the use of surnames.
WIll that be the case for you? No idea. Think of it as an investment that will let you know instantly whether someone is in youe line of descent, or part of your branch, or not, from this point forward. The value comes from knowing how to use the information you get in relation to other close testers.
I would search for males with your surname at FTDNA and see if any of themhave tested BigY and what the haplogroup is...and how many kits there are... Then find where that SNP is on the Big Tree...how old it is estimated to be etc...to judge whether a match would be useful to you or not if you happen to match that kit...
Bottomline, someone has to start testing so there is something to compare against...so, we're back to investment again...:)
If you just want to know if you're European or not...stick with 23andMe... If you want your Terminal SNP (endpoint) that defines your direct male line from you all the way back...do the BigY.

Comments

There's a cost perspective here too. I did the Big Y, and I had help funding it. Then, I added a separate autosomal test (because that wasn't included in the Big Y). So, all told, that was somewhere around $500 US that got spent (not all by me though).

If I had known then what I know now about how the different companies cover different portions of the chromosomes and therefore comparability is complicated, I might have opted to go for Whole Genome testing instead because it would only be another $500 to do that. I don't have the extra $1k in my budget right now to do that, but that may be a future endeavor down the line to help with my maternal side.

---

True, although Whole Genome testing has it's issues as well...mostly no one to compare to...:) Many seem to end up uploading to FTDNA and paying the extra fee for access to comparisons...If you went that way I'd skip BigY... FTDNA doesn't test in those areas because they haven't shown (yet) to have any genealogical relevance. But now that their new owners are wanting to move into the health aspects of it all like 23andMe maybe we'll see something similar down the road...I'd probably contact FTDNA customer Service first and make sure there's no issues in transferring etc before I went that way...no surprises! lol  BigY gives you around 700 STRs...and yet no one has really looked at anything over y111 being genealogically significant. Although I have heard a a couple isolated instances of something being ID'd ...I think it was among the Stewarts...hm.

---

BigY-700 is only $349 or so now (vs. a $1k cost 8 years ago)...  I'd wait until whole-genome testing costs the same before I'd recommend that over a BigY (NGS test)...  Just sayin'

Answer 8

The Family Tree Y-DNA test (Direct Male Line) at the 64 marker and later 111 marker range helped me with some interesting loose ends.

1. It gave me a list of my CLARK DNA cousins who had family trees which broke away from mine back in the early 1700s.  Thus I was able to find out all sorts of information I had been missing.

2. I found a couple CLARK DNA cousins who came from lines of babies who had been adopted for one reason or another.  They did not know about their Clark bloods lines prior to this.

3. I discovered a few DNA cousins who do not have the CLARK last name.   But they are Clark descendants.  After a couple years of pieces together information it turns out that they descend from families who owned plantations near the Clark's back in the early 1700s in old Virginia.  It would seem that one of the Clark boys was spreading the love around to any gal within one day's riding distance.  Now we know why he was cut out of his fathers will. 

4. I also discovered that I have a few DNA cousins with the last name Hurst and Hirst.  An old place name from Northern Yorkshire and Southern Scotland.   The links to this name predate any known Clark information  ( Late 1600s ) and probably even further back.  I even found two DNA cousins who still live in the UK.   So... either Hurst was the original name until somebody " became a CLARK" by occupation, or Some Clarks from Scotland injected their DNA into the Hurst line back during one of the various invasions that went back and forth. 

Answer 9

Echoing what Barry Smith said, "Any test can either prove useful, or not useful, and you don’t know until you do it. Even if you don’t find immediate use, if you wait long enough things might change."

For me, the FTDNA BIGY500 test seemed a complete waste of money until a couple of years later when I was able to piece together enough information to overcome an NPE.  Now I am upgraing to the BIGy700

Answer 10

I am not a DNA expert but will share my mixed experience.

1) Usefulness is mainly contingent upon the quantity and quality of your specific YDna matches. Since you can't see that information ahead of time, it is impossible to predict.

2) Had some near "groundbreaking" discoveries. A) Confirmed that a distant cousin of an adopted father was part of my Surname ancestry and was able to focus in and discover the cousins' grandfather for him - groundbreaking for him - not so much for me but very rewarding. B) My known surname paternal ancestral tree only went back to about 1800 NY - all speculation after that. My YDna matched to several "known" descendants of a branch from 1600 's Kent England that migrated to Connecticut. I was the first of the NY Beers Branch to YDNA test and confirm the connection to the Beers of Kent England. Didn't tell me exactly the common ancestor nor fill in the gap from the common ancestor to my oldest known ancestor but help narrow the focus of the research. Again this depends completely on your matches and their "known" trees. Also was able to determine from this that mom and dad were approximately 11th cousins descended from the Kent England Beers's

3) Did my testing on FTDA. The really low testing levels mainly proved worthless, and few had associated trees. Very Very few matches at the lower levels shared my surname (perhaps a huge number of adoption and name changes but not likely)  Personally recommend the mid to upper price level testing (like the 111 marker test) unless you are a bit of a DNA expert. Hope this helps. For me the 37, 67 and 111 markers were usefull, anythng below that was pretty useless - you results might be a bit different.

Answer 11

The usefulness of Y-DNA testing is dependent on many factors, including family origins, accuracy and completeness of one's family tree, the accuracy and completeness of others' trees, etc.  For me, Y DNA testing has been critical. My father's family can be traced back to early 18th century in Pennsylvania, Virginia, and North Carolina, prior to moving on to Tennessee and other areas. I took a Y DNA test when it first became available publicly and later took an autosomal test with Ancestry. Ancestry remains my preferred provider because of the sizes of both their DNA and records databases. I do use several other providers, for example MyHeritage for their DNA segments analysis. The size of the Ancestry DNA database, coupled with the chronology and geography of my family origins, led to a number of false positives in my autosomal data. I identified the false positives, or ambiguous results,  when I  discovered alternate bloodlines while analyzing autosomal data and family trees.   Over time, my Y-DNA results have become increasingly valuable for identifying and confirming  my surname line. For example, I had found autosomal matches pointing to two other well-known George lines, which I have been able to eliminate through a comparison of Y-DNA test results. My greatest success came through the use of both autosomal and Y-DNA testing. Several years ago, I found a growing number of autosomal matches that converged on Edward George ( - 1737), a Quaker who had removed from Llangurig, Montgomeryshire, Wales, to Radnor, Pennsylvania, about 1710. I order to verify the autosomal results, a relative of one of my autosomal matches took a Y-DNA test and we turned out to be a close match. We have since, along with others, taken a Big Y 700 test, which has further confirmed our line, Lineage 2 in FTDNA's George Y-DNA Surname Project. We do not yet have enough Big Y testers in Lineage 2 to identify branching in detail, but some other lineage groups with more testers are getting good results. , In my situation, the use of Y DNA tests in conjunction with autosomal testing and traditional research has proven to be a key factor in braking down brick walls.

Unless one is absolutely sure of their surname line, starting out with a Y-37 test would probably be a good way to proceed. The results of a Y-37 would provide a good basis for future planning.

Answer 12

It is a roll of the dice.  I started my Y-DNA trek in 2004 with a 37 marker test at FTDNA that came up with 0 matches for my haplogroup, E-M35. This was the best 23andMe could do also at the time.  Upgraded to 67 markers and still 0 matches.  So.....

After many years of diligently building out and verifying family lines all claiming to descend from "my" immigrant ancestor in 1636 from England to Mass. - I now have found four other men of my surname, not including my brother or myself, who are proven to be true matches -  two via STR testing only and two + brother and myself via both STR and Y-700 SNP matching.  Today, still, none of us have any surname matches   During this same time period I corresponded with and fronted a 67 marker test for a same surname fellow living in England who claimed to be of the same ancestral lineage but he turned out to be of the "I" haplogroup category - so definitely not a match!

At this point our confirmed MRCA (most common recent ancestor) is proven back to the eldest son, born 1641 in the Boston area, to "our" immigrant ancestor. 

From my perspective Y-DNA testing is well worthwhile but you can tell from some other comments - your results may differ.  So carefully consider what your objectives are before undertaking this journey.

Now, the initial 35,000 year old M-35 haplogroup origin projection has been transformed into an estimate of less than 100 years (from SNP Tracker tool) for me and my brother. Our father was born in 1901.

The cost for all this was pretty significant, but now, at nearly 90, I am quite pleased and satisfied with the results.