What Does and Doesn’t a Y DNA Match Mean?

It’s easy to forget how foreign this landscape looks to a newbie, but the newbies are our next generation genealogists and genetic genealogists.

This week, someone e-mailed me who had tested at Family Tree DNA and asked how to contact their Y DNA match they had found in a project that I manage.  I thought that was a very strange request, since your matches are on your personal page along with their e-mail addresses, so I asked for the name on their kit and their kit number so I could take a look.

As it turns out, they had no Y DNA matches on their personal page, so they were hunting for matches elsewhere.  They had joined the haplogroup E1b1a-M2 project and it’s there that they found their “matches” that they were asking about.  I commend their tenacity in hunting for matches and finding them in a project, even though they weren’t exactly what they thought.

The kit number here is 343629, Lewis.  You can see in the screen shot from the haplogroup E-M2 project page that they don’t match anyone exactly at 12 markers, and their closest match is to Harris above their entry, and they have 3 mismatches at 12 markers.

y match 1

As it turns out, Lewis and Harris didn’t qualify as matches, which is why they weren’t displayed on their personal match page.  This explains why kit 343629 was asking me how to contact their “matches.”

Family Tree DNA has set up match thresholds.  For someone to be listed as your match, they need to have no more than the following total number of mutations difference from your results.

Markers in Panel Tested Maximum Number of Mutations Allowed
12 0 unless in a common project, then 1
25 2
37 4
67 7
111 10

The reason for these thresholds is that DNA mutates at an “average” rate and for someone to have more than this number of mutations in that marker range means, generally, that the match is too far back in time to be genealogically relevant.  For people who do have matches, you can utilize Family Tree DNA’s TIP calculator to obtain an estimate of how distant the most recent common ancestor (MRCA) might be from you and your match.  I wrote about the TIP calculator and the MRCA both, so refer to those articles for more into on those tools.

The next question this person asked is, “How do you tell which markers indicate someone as a close cousin?”

The answer really involves several aspects or DNA testing, and I’m going to answer their question here, in pieces, so that everyone can benefit.

  1. In general, finding family via the Y markers is not about “which markers” as much as it is about the number of matching markers. If you share a common ancestor, the DNA of that man’s descendants will accrue mutations over time. If the common ancestor is before the advent of surnames in the culture in which they resided, then the surnames may not match, but the common ancestor still existed.
  2. In general, 12 markers is not sufficient to determine a common ancestor, although you can rule out common ancestors in a genealogical timeframe, generally accepted to be 500-800 years, by high numbers of mismatches caused by mutations. I would suggest this person test at higher markers because sometimes people do pick up matches at higher levels where more mutations are allowed, especially if the mutations happened, for some reason, in the lower panels but few happened in the higher panels. I do see this when writing the Personalized DNA Reports for people, not often, but it does occur, especially at 111 markers.
  3. You cannot necessarily identify a “close cousin” or any specific relationship utilizing Y DNA testing alone, especially at low marker levels, such as 12 and 25.  Although if someone matches you on all 111 markers, there is a very good chance that you share a common ancestors in just a few generations. What the traditional Y test (meaning not the Big Y test) does confirm is whether or not you share a common paternal ancestor and then it’s up to genealogy and autosomal testing to determine how close that relationship might be. The number of matching Y markers can provide hints and generalities through the TIP tool, but nothing more.
  4. For this individual, in addition to upgrading beyond 12 markers, I would recommend that they take the Family Finder autosomal test because that will provide them with a list of cousins on all of their lines, not just their Y line. Based on their earlier commentary, they are looking for all family, not just their paternal line. If you have Y matches and autosomal matches, through the Advanced Matching tool on your Personal Page you can see who, if anyone, is a match to you on both.
  5. However, all of this said, the combined pattern of Y markers, not individual markers, determine the match or non-match, and it is your personal DNA signature. Think of it as a song and the markers as notes in your own personal DNA song. Given that mutations arise in each person’s line, sometimes the various DNA mutations are rare, and those rare markers together can be utilized to determine how closely one might match someone else, especially if the surnames don’t match. I see this often in African American descendants of slaves because surnames weren’t adopted until after the Civil War ended in 1865. Often the 1870 census is our first opportunity to find these families with a surname, and sometimes they subsequently changed their surname.

One of the things I do for my customers as part of a Personalized DNA Report is to complete a profile for them of the relative rarity of their DNA by marker.  Please note that I don’t do DNA reports for people who haven’t tested at least 37 markers because I don’t have enough information to work with.

In the case of this individual, I compared their 12 markers in my database of haplogroup marker frequency with the following results.

y match 2

Values under 25% are bolded, as they are rarer values and the combination of these rarer values are likely to be your own personal family line rare marker DNA signature.  Said differently, you are more likely to be more closely related to those who carry this rare marker signature than those who don’t.

This person has 6 out of 12 markers that are relatively rare.  Normally, one would expect no more than 3, so this is likely why they have no matches.  This is a good news, bad news thing.  The bad news – no matches today.  The good news is that these rare markers value, combined, are a wonderful personal filter that eliminates matches by convergence.  So, someday, when they do have a solid match, it will be relevant and not just because they have all common markers.

And now for the next question.  How can you obtain your own list of marker frequencies?  Obviously, you can order the DNA Report for $349, or if all you want is the marker frequencies, you can order a Quick Consult for $50 and can obtain all 111 of the Y marker frequencies for any one kit.

Guarantees

Most people just want an answer.  I fully understand that.  Me too, but often, that’s not how DNA testing and genetic genealogy as a whole works.  So the question, “What test can I take to give me the answer?” really doesn’t have a solid, works every time, answer.  There is no absolute, no guarantee.  Sometimes, depending on the question at hand, a regular Y DNA test will do exactly what you want.  Other times, like in this case, not so much.  But you won’t know until you test and there is no way to predict an outcome.  Testing may provide the answer in spades, immediately, and it diggingdoes sometimes.  Other times, you get a puzzle piece with a fortune cookie note that says ”you will undergo more DNA testing.” The answers are tied to DNA testing, yours and other peoples, traditional genealogy research and sometimes, luck.  But it has been my experience that those who work the hardest, test most thoroughly and dig the deepest are most often the ones who experience more occurrences of “luck.”  Keep digging.

As Louis Pasteur said, “Fortune favors the prepared mind.”  Not nearly as eloquent as Dr. Pasteur, my old Hoosier farmer Dad would have said, “apply a little more elbow grease.”

I hope this has helped to clarify what a Y DNA match actually does and doesn’t mean, and how to take the next step in finding your family.

44 thoughts on “What Does and Doesn’t a Y DNA Match Mean?

  1. Roberta, are these frequencies something you have calculated? Do you have a frequency database for STRs in other haplogroups as well? What a great analysis tool!

  2. Roberta,
    In point 3 above you say “Although if someone matches you on all 111 markers, there is a very good chance that you share a common ancestors in just a few generations.” In one of the DNA projects that I’m an administrator for I have two men who match exactly at 111 markers yet each has taken Family Finder and have no matches in common there. Isn’t that kind of strange?

  3. Hi Roberta-Speaking of yDNA implications, I would think that if I match someone at 67 of 67 I might reasonably expect that we both would have some autosomal sharing too, but right now that seems not to be the case. We have both tested autosomes at FTDNA and 23andMe, but at the default level of both services we have no matches on either service. (I have not been able to upload to GEDmatch yet.) I know that my settings on 23andMe are open enough that my daughter showed up matching me appropriately. My son, however, does not show up as an autosomal match to me on 23and Me. I’m pretty certain that he has simply not granted the correct permissions, though he claims he has. I’m wondering whether my 67-67 (FTDNA) y-match has missed a permission gate also, or whether 67-of 67 on the Y is not a strong enough indicator to make some autosomal sharing a virtual certainty? Hoping you will be speaking in our area again soon.

    • I have seen a number of 67-67 matches that don’t share autosomal. Remember that autosomal only reaches back at the threshold levels used by the companies 5 or 6 generations and it is certainly possible to have a 67-67 further back than that. I have had better luck finding smaller matches at GedMatch for those people. It does tell you though that your common ancestor is quite a ways back in time. Regarding your son and your files. Download them both to GedMatch.

  4. Roberta,
    Thank you for this explanation regarding 12 Marker Y-DNA testing.

    My Dad tested at this level a few years ago but I have only recently begun trying to make sense of the results. It appears the 12 Marker tests are inadequate for discerning genealogical information. A Family Finder DNA test was run and those results have just recently been received. At this time, he is identified as hapIogroup I -M253. Previously it had been I-M170.

    There is one match (12/12) for Dad at the 12 marker level. Other than attempting to contact the match, I’m not sure what else to do with the information we have. I will re-read and study your explanation here as it exactly addresses my situation, and appreciate your time in writing it up.

  5. Roberta,
    Another very informative article. Question for those who may be considering it… If someone has already done the Family Finder Autosomal test at FTDNA and then wishes to do the Y-NDA 111 test, do they have to go through the whole cheek-scraping thing all over again? Or does FTDNA just pull their autosomal test sample off the shelf and test that for Y-DNA?
    Thanks, Tom

  6. Another great blog entry. Thanks, Roberta! I have had my brother’s DNA tested to 67 markers, we’ve done autosomal testing, full mT testing, and the D9S919 (per your blog’s suggestion for possible Native American ancestry. It gave us the result of 15-17, so no luck.) Our main goal is to find out why our family was listed in census records as “mulatto.” It is a mystery because their last name is Conner. Civil War and WWI records indicate they all had black hair, light skin, and blue eyes (my great grandfather served aboard three Union supply ships during the Civil War and is listed as mulatto in those records, though it also says he had blue eyes.) They’ve owned land at least since the early 1800s in an area that used to be predominately Tuscarora Indian. They changed their name from Conner to Carney after moving out of their native Jones County, North Carolina and into Duplin and Pender counties.
    My first thought was a white/black/Native American mix, but autosomal testing indicated 97% European (Western Europe, Central Europe, Scandinavia, and Southern Europe), 3% Central South Asian (Afghanistan area), and 1% Eastern Middle East. From that, I can kind of see a possible migration pattern from the Fertile Cresent up along the edge of the N. Mediteranean, to Europe.
    I feel that my question is a simple one to ask, and yet I don’t know what else to do to find the answer. My brother, father, and many of my father’s cousins look Hispanic or Native American, but with gray or blue eyes. (My brother once grew a mustache and several Hispanics started speaking Spanish to him on the street!) I look totally Irish, as does my sister.
    Any suggestions that might also help others in the same boat?

  7. Roberta, this one’s a keeper for the project I administer. Thanks. We are I1 (Big-Y pending). Ken Nordtvedt’s I1 YDNA Haplogroup Project says that “DYS455=8 is a very good predictor” to tell if you’re an I1. Is it possible that an I1 SNP can be found within the STR marker DYS455? Or, do SNPs have to fall outside STR markers? The spreadsheet for Ken’s I1 Modalities research displays a row/column relationship between SNPs and STR markers. This is not a beginner’s question, I realize, but sooner or later, they all want to know more about SNPs and STRs.

      • Roberta, I know you’re very busy! Thanks for the reply. I re-read your 2014-02-10 STRs vs SNPs article including all the comments. I get your comment that “First of all, you can’t compare SNPs and STRs. They are two different things.” (Re: May 25th comment to Richard McMurtry). I also get it that Dr. David Mittelman declares “There is no such thing as a SNP/STR.” I agree with you both. I am NOT trying to compare SNPs and STRs. I am not trying to declare that there IS such a thing as a SNP/STR. What I am asking is whether a SNP can exist INSIDE an STR? Yes or No?

        If the answer is yes, that tells me a SNP could alter the usual number of STR repeats. If the answer is no, that tells me a relationship that may appear to exist between a SNP and an STR is either pure coincidence or there’s something else present that affects them both.

      • I asked Dr. Mittelman how this would be handled if a repeat sequence had a SNP in the middle of it. Dr. Mittelman’s reply is below:
        A SNP can certainly exist in a repeat tract. You could have something like this:

        CAGCAGCAGCAG (4 CAGs)

        CAGCAGCACCAG (4 CAGs, but the third triplet has G mutated to C)

        I would still describe both repeats at CAG(4). It is true that the first one is a perfect repeat, and the second is not 100% pure repetitions. I believe at the nucleotide level this happens for at least some STRs, they are not always perfect repetitions.

      • Thank you SOOO much for the priceless info both you and Dr. Mittelman provided. I’ve searched the net through countless articles looking for this. This explanation is not to be found anywhere else. It has helped clarify questions had by two projects that I know of. You can bet I’ll be quick to share this. Thank you very much and please thank Dr. Mittelman for us. ~Ron.V

  8. Roberta,

    Thanks for the great post again. We have tested two paternal donors with FTDNA at the 37 marker level. We felt we were certain that we had a common ancestor back 6 generations. BTW we could never authentic that , so we resorted to DNA. Of course the results came back and we are not a match. Now I understand that they are haplogroup R M269 and if you go to ysearch.org they will give you a SNP for RM 269, R1b1a2. I searched on ysearch for R1b1a2 and it showed still no matches. I am not sure if I am doing it right. My question is basically, where can I go from here ? The BRICK wall is there, but I need more assurance I have exhausted all possibilities.

    Thank you in advance

    Elaine Gaudet 9 Marjorie Cres Stratford, PE. Canada C1B1X3 Sent from my iPad

    >

  9. I remember what a thrill it was, back in 2006, to receive my first 12 STR results from the Genographic Project. I thought I matched one of the Highland clans! That was indeed ancient DNA testing, in more ways than one. Four of my 26 matches at 12 (which includes up to a genetic distance of one) have R-L21 as their SNPs, which probably split off from R-P312 about four thousand years ago. My branch of Haplogroup R is way down on a little twig from R-P312. And as I have no matches below FTDNA’s cutoffs at 37 nor 67, I’ve given up on Y-STRs being in any way helpful genealogically. Maybe one day, sigh. The best bang for the buck genealogically, IMHO, is Family Finder and other autosomal tests. It’s enabled me, as an adoptee, to match up with cousins I never knew I had. Thanks for another superb post.

  10. Does the term “mutation,” as used below in describing FTDNA’s matching criteria, refer to the number of differing markers without regard to the number of steps for which a marker differs? (In other words, if two men differ on only one marker, but by two steps on that marker, is this one “mutation” or two?) __________________________

  11. Hi Roberta,
    This post is great, and a bit of an eye-opener. The issue about not sharing autosomal DNA with a Y-DNA match has been very puzzling to me. My great grandfather was born out of wedlock in 1870s. He took his mother’s maiden name; TAYLOR, but his mother told him the last name of his father, so his paternal surname was “known” within the family for generations. We just had that last name confirmed with the FTDNA Y-DNA37 with my 91 year old father’s DNA. I have been in contact with my father’s matches, (36/37 and 35/37 matches) and have narrowed down the men with the same (newly confirmed, uncommon) surname who lived in the small county in which my great grandfather was born.

    What is peculiar, is that my great grandfather’s mother (my g.g. grandmother Taylor) was crippled and couldn’t have traveled far to meet a suitor, so the father would have had to be within close proximity to her. There is science (DNA) and then there’s geography. I cannot get a match with family finder with any of the ancestors of any of these men. I’ve also searched Gedmatch.com. and compared my father’s DNA with known ancestors of the men in the county, especially ancestors of those who men lived close to her.

    Would we benefit from a personalized DNA report, or has the right descendant on the paternal side not been DNA tested yet?

    Thank you in advance for your response.

  12. I am here from Yvette’s blog, and I am so glad to see some cogent explanation of some things! Could you help me confirm what I *think* I understand from my own Y DNA data at FTDNA? There are a number of prominent and well-documented early U.S. families with my father’s surname (Kauffman/Coffman), but my own research hit a brick wall before I could link into any of them. Using my uncle’s Y DNA (my father is deceased, and I am female) I joined the surname project at FTDNA, hoping for a match with one of the well documented families. If I am reading the project report correctly, I do NOT match the family lines represented there, with one exception–I may match one other person who also doesn’t know an ancestor earlier than about 1800 or from a country other than the U.S. Do I understand this right? My kit is N41261 and hers is 101991. I can send you a screen shot of the surname project page.

  13. Pingback: DNAeXplain Archives – Introductory DNA | DNAeXplained – Genetic Genealogy

  14. Hi
    I have a basic question regarding the result of alleles of STR 111 marker test. The comparison with someone else is that should be within a certain Haplogroup or possibly to compare my results even randomly with someone else in other Haplogroup. Second thing, regarding the understanding that my 111 markers and my brother 111 markers must be identical, so May I know what is the ratio of the identical numbers of 111 marker results between me and my father, as well as between me and my grand father then father of my-grand father and so forth. Is there a certain percentage consequent with the chronological temporal line? Similarly, who will shows more markers similarity my cousin or my grandfather? Kindly let me know if there is any comprehensive way of calculation for STR 111 marker. Regards

    • Family Tree DNA compared your results with anyone you match, but it will very likely be within the same base haplogroup. Mutations can happen any time, so your markers don’t have to be the same as your brothers or fathers. I don’t understand the question you are asking, but the TIP calculator provides estimates to the time of most recent common ancestor.

    • Ash,

      Roberta said she doesn’t understand the question you are asking because you may be confusing results analysis of yDNA with autosomal DNA? yDNA markers typically don’t change from generation to generation. The “ratio of identical markers” for yDNA between you and your father or grandfather or you and your cousin (who shares the same grandfather in your male line) will usually be identical, that is 100% or a ratio of 111/111 for the y-111 level unless there is a mutation.

      On the other hand, autosomal DNA differences are great from generation to generation. For example, you share only half the autosomal DNA or your parent, 1/4th of your grandparent, 1/8th of your 1st cousin, and so forth (see: http://www.isogg.org/wiki/Autosomal_DNA_statistics).

      • Hi Ron. V
        Thanks for the explanation. I saw some results of 111 markers in many project in Family tree DNA being classified as samples descended from the same great grand father. Rather, those sample’s have results are not shows 111/111 which some variations in some marker numbers appear amongst them. For this reason I thought those who are living at closer temporal distance have closer result of 111 markers and for example the 111 markers rests of my great grand father from 1000 years ago will not be 111/111 with mine due to the big temporal gap between me and him. I think you explained that and you said the 111 markers must be identical unless there is mutation. Thus, the question here what make the new mutation emerge and is it a common case?
        Regards
        Ash

      • Ash, you asked “what make the new mutation emerge and is it a common case?” Mutations emerge at random but can be identified through research. Research is statistical and therefore general. It’s like saying, “Odds are, you’re genetically [a certain nationality or ethnicity].” Odds are those in my I-Z63 Haplogroup have a yDNA mutation every 3.3 generations according to the latest research. What makes a mutation emerge? Mutations are a normal part of genetics. They happen all the time. Here’s a link that helps explain it:
        http://learn.genetics.utah.edu/content/variation/mutation/
        The question is not whether they emerge but whether they’re passed on to the next generation. They emerge all the time but are rarely passed on through yDNA, or mtDNA. The differences in autosomal DNA are just that — differences, not mutations. We speak of mutations only in yDNA or mtDNA. I don’t know much about the X chromosome so you’ll have to ask someone else about that. Is there anything common that causes mutations to be passed on? Not that I’m aware of. They happen at random and are passed on at random. Research can only identify how often they occur. Again, that research is VERY specific depending on your Haplogroup. Keep in mind that this type research is also VERY new and new discoveries are happening every day.
        Ron.V

  15. Hi Roberta
    I am trying to figure out my Pratt ancestry. I know my US patriarch immigrated about 1630 from England. I have read that the first Pratts to come to England came over after William the Conqueror and were of the Norman family Pratellis. Pratellis was anglicized into Pratt/Prate. If many of the English Pratts came from this Pratellis family, would testing modern day English Pratts be useful. I have taken a FT Y-DNA 111 test. In other words, after a thousand years, what are the chances that testing a currently living English Pratt with the YDNA 111 would yield anything useful? I have offered to pay for such a test with an English Pratt but don’t want to waste my money if it is unlikely to yield anything meaningful.
    Thank you for your time and consideration!
    David Pratt, M.D.
    BTW: great web page here where you explain things in a cogent manner.

    • There is no way to know if you don’t test. I would pay for that test in a heartbeat, if I were in your position. Even if you don’t match at 111, the Big Y will tell you whether you match before that or not. My Estes line matches the English Estes line that diverged in the 1500s on STR markers.

  16. The Family Tree DNA website says a match at 12 markers indicates a 50% chance of MRCA within 7 generations, and a 95% chance of MRCA no more than 29 generations ago. But you are saying that is hooey? Thanks. I wish I would have read your page before I paid for my test.

    • I didn’t say that’s hooey. Those percentages may be right, especially if you’re working with the same surname. What I said, the long way, was that DNA analysis is a process. Sometimes you’re lucky. Sometimes you’re not. Fortunately, there are more tools available to help us obtain our answers.

  17. Just starting to puzzle out what a y-dna match means. I did the 111 test, and at the 25 level I have a couple of matches with women. Is that possible or is it just that someone has labeled a male dna swap with a female name?

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