Wednesday, February 24, 2016

How Could DNA Testing Help Us?

I mentioned in a previous post that I had hopes that DNA testing could help us solve the mystery of Nicholas Ackley's ancestors. Now is as good a time as any to discuss this, so here goes. This is a pretty lengthy post because there is a lot to cover, and we will only scratch the surface. My goal in presenting this information is to try to get some momentum behind testing Nicholas Ackley's descendants and sharing results so we can try to make some progress in finding/verifying his ancestors.

I am relatively new to DNA testing, so I've tried to get up to speed on it by reading about it and subscribing to several blogs that discuss various aspects of using DNA in genealogy. I still have a lot to learn, but I think I have enough of a handle on it to have some ideas on how we could come together as a group and use it to our advantage.

If you are considering DNA testing, there are a couple of decisions to be made: which company to test with, and which type(s) of tests to do, which we'll discuss next. If you've already tested, hang in there, we'll talk about what we can all do a little later in this post. 

Testing Companies

Although there are many companies that do DNA testing, there are really only three that you should consider for genealogical purposes. I regularly read a blog called "DNAeXplained" written by Roberta Estes that I think is worth reading if you are interested in using DNA for genealogy, and in this post she gives some good reasons why you shouldn't look anywhere but the "Big 3". As she writes, the "Big 3" are Family Tree DNA (sometimes abbreviated as FTDNA), Ancestry, and 23andMe. Roberta points out in her post that 23andMe is going through some upheaval right now as it relates to their genetic genealogy products, so for the time being she is recommending against 23andMe. So, that leaves FTDNA and Ancestry. I have experience with both companies; I have tested myself as well as seven other family members with Ancestry, and I have also transferred my results to FTDNA (for a fee) for reasons I will discuss below.

Types of Tests

There are four types of tests, each of which can be helpful for genealogy purposes in different ways. Roberta has a much more detailed post about the types (Y-DNA, mitochondrial, autosomal, and X) here, but for most of us autosomal and Y are the tests we are most likely to do, so I'll stick to these two here:

Autosomal DNA

Autosomal DNA is the DNA inherited through the autosomal chromosomes. Humans have 23 pairs of chromosomes; the 22 numbered pairs (1-22) are the autosomal chromosomes, and the 23rd pair are the sex (X and Y) chromosomes. Both parents contribute autosomal chromosomes, and the results of autosomal testing, along with genealogical information, can be used to discover familial relationships. Stated simply, you can identify your parents, siblings, aunts, uncles, cousins (recent and distant), and so forth using autosomal DNA test results. Both males and females can do autosomal testing. Both Ancestry and FTDNA offer autosomal testing; each has strengths and weaknesses when it comes to the tools provided to analyze your results. I'll talk more about that in the "What Do You Get When You Test?" section below.


The Y chromosome is male-specific; the pair of sex chromosomes of males is made up of an X contributed by the mother and a Y contributed by the father, while females have two X chromosomes, one each from the mother and father. As a result, only men can do a Y test (sorry ladies!). However, all is not lost, women can get Y information for their family line by having their father, brother, uncle, or cousin in the direct family line test. The Y chromosome is passed from father to son virtually unchanged with a predictable rate of mutation, so the Y-DNA test is typically used to determine if there is a shared ancestor within a genealogical time frame. This is typically the test used in surname projects. Currently only FTDNA performs Y-DNA tests, so if you are interested in doing a Y test that is your only choice. Ancestry used to do Y tests (I did one several years ago before they stopped); if you have Y results from Ancestry you can transfer them to FTDNA for a fee and take advantage of the tools they have (which I have just done - waiting for the final results).

What Do You Get When You Test?

Both Ancestry and FTDNA have mostly similar tools for autosomal tests, but each has what I see as a major weakness that the other one just does better. We'll get into that shortly, but here are the things they probably do equally well.

At the highest level, they both offer an estimate of your ethnicity. In fact, that is what Ancestry calls it, your ethnicity estimate. Here is what mine looks like:

Ancestry Ethnicity Estimate

You can see that it splits out the various parts of the world your ancestors likely came from along with the percentages for each. FTDNA has something similar that they call myOrigins. Here is mine:

FTDNA myOrigins

You can see that at a high level the FTDNA estimate is similar to the Ancestry estimate, but what is a little surprising to me is that there is a pretty big difference between the British Isles estimates (51% vs. 29%) and the Southern Europe estimates (about 4% vs. 20%) between the two tests. Apparently this is not a real exact science; it is highly dependent on the algorithms and reference groups used by each company. Ancestry has a good explanation of their methodology on their website; they basically compare your DNA to the DNA of people who are native to 26 global regions they have defined. FTDNA doesn't give a real detailed explanation of what they do, but it sounds like their methodology is similar - they compare your sample to reference populations from around the world.

At the next level of detail, Ancestry and FTDNA are also pretty similar. They both offer testers a list of the people they match with based on thresholds for the lengths of matching segments. The methodology each uses is different and results in differing numbers of matches, but ultimately they both give you a list of your matches, an estimate of the strength of the relationship; e.g., 3rd cousin, 4th cousin, distant cousin, etc., and a way to contact your matches. Ancestry has an internal messaging system that allows you to send notes back and forth with your matches, while FTDNA gives you the e-mail addresses of your matches so you can communicate with them directly. Here is a portion of my match list from Ancestry (note that I have masked the names of my matches):

Ancestry Matches

Ancestry segregates your matches by strength of the relationship - the example above comes from my "Distant Cousin" list, which is anything from 5th to 8th cousins. Note that there is an indication as to whether the match has a family tree, and if so how many people are in the tree. The existence of a good sized tree is important in trying to determine how you and your match might be related; in these three examples I won't have much luck. The first person has no tree, the second one only has 2 people, and the third one has a little lock next to it, meaning the tree is private and I can't look at it. I consider Ancestry's huge database of trees to be a major advantage over FTDNA, but unfortunately the situation shown here happens more often than I'd like; in such cases you can't really figure out how you're connected.

FTDNA has a similar approach to presenting detailed results. My list of FTDNA matches looks like this (once again I have blanked out the names):

FTDNA Matches

The little icons next to the silhouettes are tools for managing each match, including e-mail, notes, family tree (if the tree is green, it means your match has uploaded a tree), and some comparison tools. The "Shared cM" column tells you the total length of the DNA segments you share with that match in centiMorgans (the standard measure of DNA).

On both Ancestry and FTDNA you can dig into more details about your matches in different ways, and this is where the two companies have the biggest differences. If I had to label them, I would say that Ancestry is more genealogy-centric, while FTDNA is more genetics-centric. Depending on what your goals are, that may be good or bad. Hopefully the next few paragraphs will illustrate what I mean.

If you click on one of your Ancestry matches, you will go to a page that looks like this:

I have erased all references to his user name and family surnames so you can't tell who this match is. There are a lot of things you can do on this page, mostly around how you and your match connect genealogically. You can see a list of the surnames you have in common in your trees and a condensed view of their tree, which you can expand by clicking on the "View Full Tree" button. Ancestry has recently added a couple of new features that are an attempt to give you more information about the DNA side of things. If you click on the little "i" icon just below the "Predicited relationship" section, a box will pop up showing the total length of your shared segments:

This is similar to the "Shared cM" information that FTDNA gives. The other new feature is the "Shared Matches" button just above the tree; clicking on that will give you a list of people that match both you and the match you are currently viewing. This is similar to one of the tools available on FTDNA called "In Common With"; this can be helpful in trying to pinpoint just how you are all connected. These two features are the extent of the tools that Ancestry has that deal with the actual DNA piece of the puzzle; the rest of their tools relate more to the genealogy piece. For example, if you click on one of the shared surnames, you will get a list from your match's tree and your tree of everyone that has that surname. By clicking on "Boettcher", this box pops up:

This is useful in trying to figure out where the connection is; you can see that we both have a Frederick Boettcher (although I spelled it Friedrich). We have actually communicated with each other and determined that Frederick/Friedrich is indeed our shared ancestor. John Carl August Boettcher (his 2nd great grandfather) and Augusta Boettcher (my 2nd great grandmother) appear to be siblings. Unfortunately neither one of us knows much beyond that, but in the future we can help each other out if we find out more.

Ancestry also has a feature they call "Shared Ancestor Hints". They find these by comparing your tree with your match's tree to see if you have common ancestors somewhere along the way. I have a few of these; here is an example of one:

The couple at the top represents the shared ancestors; the two columns of people below them are the line of descent to me and my match to show how we are related. This match is my 8th cousin. This feature only works well if you and your match have well-developed trees and you happen to have spelled the names all the same (a different spelling of Frederick in the last example is why that match and I didn't have a "Shared Ancestor Hint"). There are a few other tools on Ancestry, but these are the main ones and it gives you an idea of what you can expect. On to FTDNA.

The genealogy tools on FTDNA are a little more crude, and as a result not as useful. For example, if you click on the little green tree icon that was pointed out above, it will show you your match's tree. It looks like this:

For me, this tool just feels really clunky. To move around in the tree you hold your left mouse button down and slide the tree around to look at different portions of it, making it very hard to really find anything. You can click on an individual and get their information, which looks like this:

That's about the extent of the genealogy related tools on FTDNA; they don't have anything like the Ancestry features discussed above. However, FTDNA does outshine Ancestry when it comes to tools designed to do analysis on the DNA matching side of things. For example, the chromosome browser allows you to compare up to five of your matches to see how you line up. Below is an example where I have compared three of my matches (whose names I have erased) that I identified with the "In Common With" tool; i.e., these three have matches to each other as well as to me.

The graphic shows how each of these three individuals matches me. You can see that they all match me on chromosome 7; the orange one has the largest segment, the blue one the next longest and most of that matching segment overlaps the orange, and finally the green one is smallest and falls completely within the end points of the orange but not the blue at all. This type of information can be used to triangulate your matches -- find groups of people who all match each other on the same segment, which is a guarantee that you are all related genetically (provided the segments are large enough, which in this case they are). The next step is to examine your trees and find common ancestors to try to establish exactly what the relationship is.

You can drill down a little deeper by clicking on the "View this data in a table" link just about the graph. This will give you a table showing exactly which segments you match. For the orange individual above, it looks like this:

You can see that we have a pretty large segment -- 36.81 cM -- that we share on chromosome 7. Anything over 7 cM is generally considered large enough to indicate a genetic relationship. Note that there are also a number of much smaller segments; these are too small to be considered significant, and can be ignored (which is why they don't show up on the graph). There are some other tools on FTDNA that allow you to look at more details, but again I think you get the idea from what is shown here.

Note that all of the tools presented above are for working with autosomal DNA tests. The tools for Y tests are different. My results for the Y test look like this:

My first match is an unusual situation - he does not have the Ackley surname, which is not normally expected (I have masked his name as well as the name of his "Most Distant Ancestor"). I'm not sure what to make of that just yet.

The other two matches both had the Ackley surname as expected (first names have been masked). Neither one of them listed their "Most Distant Ancestor", which is not very helpful. The "Most Distant Ancestor" is the type of information that could help in connecting otherwise unconnected lines of a particular surname. For example, if I had a match with someone who listed someone other than Nicholas Ackley as "Most Distant Ancestor", it would behoove us to look at each other's genealogies to see if we could determine a common ancestor. This could result in either an extension beyond Nicholas for me or a connection for my match to the Nicholas Ackley line that was previously unknown.

There are a couple of other things to take note of in this report. The "Genetic Distance" column is a measure of how close we match. A genetic distance of 0 means that we match each other exactly on all 12 markers tested. A genetic distance of 1 means that there is one marker where our values differ by 1. A note about the number of markers -- my original Y test with Ancestry was a 33-marker test, but not all of the markers were the same as the markers that FTDNA uses. So when I transferred my results to FTDNA, I received a test kit so I could give them a sample so additional markers could be tested. Until that kit is processed, I only have 12 markers that can be used for comparisons.

Note that next to each match there are a series of buttons that can be clicked to get additional information or send e-mail. Clicking the orange "TiP" button gives an additional report that looks like this:

"TiP" stands for "Time Predictor" and gives an indication of the probability that two matches have a common ancestor within the given number of generations. The more markers you test, the more accurate these calculations become.

A Couple of DNA Success Stories

OK, so how do you use this stuff in practice? I have had a couple of experiences that I think can illustrate how this could work. The first is from my mother's side of the family. My maternal grandmother's maiden name is Jones -- about as common as it gets (#5 on that list of surnames from the 2000 census that was discussed a few posts ago), which creates some problems -- there are just so many of them! Her father was born in Chicago, and after a little digging, I was able to figure out that her grandfather's name was William Jones. Well, there were over 50 men named William Jones in the Chicago directories, so that wasn't much help. I had kind of resigned myself to the fact that this might be a brick wall that I wasn't going to break down easily.

But then DNA testing came along, and I decided to have several family members tested, including my mother's cousin from the Jones side of the family (my mother was an only child and passed away before I could get her tested, so the closest living person I could get was her cousin). Shortly after my cousin's results were in, a match showed up whose last name was Jones, and boy did he look like my great uncle! I contacted him, and we began to compare notes, and soon discovered that we had to come from the same Jones family. We both had to do some more research to make the final connection, but in the end we were able to conclude that our Jones families came from England to Chicago in the 1860s and started a brick manufacturing business together. The DNA match led us to each other, but ultimately it was the diligent genealogy research we had both done that allowed us to connect all of the dots.

I have another DNA story that is still being written, but it will illustrate the point. Another brick wall on my Ackley side of the family is my 2nd great grandmother Marie (Jeannerette) Ackley. The story on how I even figured out her maiden name was Jeannerette is a topic all on its own; the problem here is that I just couldn't find anything on her father. The only clue I had came from her death certificate and the death certificate of a man who I'm reasonably certain is her brother. They both listed their mother as Mary Cecil, but one said the father's name was George and the other said James. Both Marie and her brother were born in Brooklyln, but I could not find any other Jeannerettes in New York. Census records for Marie and her brother were not consistent; some said their father was born in France (as did their death certificates), but others said he was born in South Carolina. In searching online for Jeannerettes I was getting a lot of hits in South and North Carolina but nothing in New York. It turns out that there were several French Huegenot Jeannerette/Jenerette families (there are other spellings too) that came to America in the 1700s, but I could not find any connection to any of them. Once again DNA is coming to the rescue. I have discovered at least three matches that all have common Jenerette ancestors from the Carolinas in their family trees. I have yet to make the final connection to them, but at least now I know where to look and I can focus my research much better.

My point in relating these experiences is that I am hoping if more Ackleys do DNA tests we might be able to start connecting with each other and solve some of the open questions we still have. There are Ackley lines in the United States that may or may not be related, and as discussed in the previous post I am not convinced that we know for sure who Nicholas Ackley's ancestors were. DNA may help us answer those questions.

Ackley Surname Project

FTDNA offers the ability to conduct projects, which are usually arranged along surname lines, but can also be geographic or ethnic in nature. There is an Ackley surname project on FTDNA (there are only 12 participants right now). Up until about 2 weeks ago the project appeared to be inactive; the link to the project website was broken and the project administrator was unreachable. I was able to get FTDNA to assign me as the project administrator, and I have managed to get the project website up and running. Here is the goal of the project as written by the previous project administrator:

Sounds like a pretty good goal to me - in fact it sounds like something I just wrote above. I am still learning about how everything works, but I think it is a worthwhile project to join if you are so inclined. If you tested with FTDNA you can join the project right now for free. If you tested with another company, you can pay a fee to transfer your results to FTDNA and then join. If you'd rather not spend the money, there is a free alternative in the next section. 


GEDmatch is a free website that allows users to upload autosomal DNA test results and offers a myriad of tools for analyzing matches. They will accept results from any of the "Big 3" companies mentioned above, and uploading is very easy. I have uploaded my test and all of the tests of my relatives. Besides the tools they offer, another big selling point for GEDmatch is that you can expand the number of matches you have because your test will be compared against people who have tested at all of the "Big 3" who have chosen to upload to GEDmatch. Here are some examples of the tools they offer:

This is a snippet of my list of matches from GEDmatch. As usual I have blanked out any real names and cut off the list of e-mail addresses in the last column on the right. Note that this is not a real slick interface, but it does what it needs to do (did I mention that it is free?). This output was produced by a utility that is called "'One-to-many' matches". You enter your kit number (this is how your test is identified) and it compares your kit to all of the other kits in the database and gives you a list of the top 1500 matches that meet the thresholds for segment length (which you can change if you want to). The top six matches in my list are my relatives who have tested; thus, the really high total cM values (EA is my dad, KA and JA are my siblings, JD is my dad's cousin on his dad's side of the family, SC is my mom's cousin on her mom's side of the family, and JL is my dad's cousin on his mom's side of the family). The last one you can see on this list is the same user who I showed in some of the Ancestry examples above -- the one with the Boettcher connection. 

You may ask why I tested so many people, and there are several reasons. I tested my dad because he has DNA that I don't have -- I only got half of my DNA from him and half from my mom, so he will match people that I don't match. I tested my siblings because they got some of my mom's DNA that I didn't get, and since I wasn't able to test my mom before she died that is important. My dad's and mom's cousins for the same reason -- they got DNA from my great grandparents that my dad, mom, siblings, and I didn't get. With the cousins you get the added benefit that for anyone that matches me and them, I automatically know which of my sets of great grandparents they are related to, and that narrows down the search. For example, the Boettcher cousin I mentioned above also matches SC (my mom's cousin from her mom's side of the family) so I knew to look only on that side for surnames that are common between us.

Another tool available is a "One-to-one' compare". You can do a one-to-one compare with anyone in your list of matches by clicking on the A link under the "Autosomal" column, or from the main page you can select the one-to-one option and enter any two kit numbers you want to compare. Here is the one-to-one comparison for me and my Boettcher relative:

You can see that we have four pretty good-sized segments that match.

GEDmatch also has an ethnicity estimator. They also use a pre-defined set of reference tests to come up with their estimates (different than Ancestry and FTDNA). Here is what mine looks like:

There are also a couple of "fun" utilities, like this one that predicts eye color. They nailed it for me:

That is not a picture of my eye; it is the picture the utility picked to best represent what it predicted for me, along with a list of rules used to make the prediction. This is really close to my real eye color, and they got it right for my dad, siblings, and wife. Useless for genealogy, but fun nonetheless.

There is one final GEDmatch tool I'd like to discuss - the triangulation tool. While most of the GEDmatch tools are free, there is a small group that you have to pay for. There is no set fee -- if you donate at least $10 you get the additional tools for a period of time. The triangulation tool is worth the donation -- it does what I discussed above systematically ("Shared Matches" at Ancestry and "In Common With" at FTDNA) and gives you a list of everyone you triangulate with. Here is a snippet of some of my triangulated matches:

What the algorithm does is compares my DNA with all possible pairs of people that match me on each chromosome and checks to see if those pairs also match each other. So every line in the above snippet is a pair of kits that match me on chromosome 15 along with the start and end points of the matching segment and the length of the match. The first pair matches me between 24,150,013 and 27,388,629, so the three of us share a common ancestor -- now we need to do our genealogy homework and figure out who it is. Note that for the next three lines, kit 1 is the same kit, and kit 2 is three different kits, meaning the person from kit 1 and all three kit 2 people match me at that same location (sort of). Note that the starting location is the same for all of them, but the ending location of the first one is a little bit less than the other two. This is OK since the last two are a little bit longer; all five of us share at least a 17.3 cM segment on chromosome 15 so we have a triangulation group of five people and we share a common ancestor.

What to Do Next

I know this is a lot of material, so you need to know what you can do next. My ultimate goal is to convince Ackleys to get together to share DNA results and help each other discover our ancestors, so hopefully I still have your interest at this point. Your next step depends on what you have already done.

If you have already tested with one of the "Big 3" and are interested in getting in on some type of Ackley project, you have a few choices:

  • The quickest, simplest, and cheapest thing to do (free) would be to upload your results to GEDmatch so you can start comparing your kit to other Ackleys who join, as well as the rest of the people in the database who might share non-Ackley ancestors with you.
  • If you tested with FTDNA and haven't already done so, you can join the Ackley project there (also free).
  • If you tested with Ancestry or 23andMe, you can pay a fee to transfer your results to FTDNA and get in on the Ackley project. The transfer fee for autosomal tests is $39, and the fee for the Y test depends; to transfer the results without any additional testing it is $19, but if you want to test some additional markers to mesh with FTDNA's standard number of markers it is $58. Once you have transferred, you can join the Ackley project for free.
If you have not already tested, here are some options:
  • Decide which type of test you want to do. Males can choose either the Y test or the autosomal test; females can only do the autosomal test. The autosomal test is more robust in that it gives you information about all of your ancestors; the Y test is restricted to your patriline (line of descent through your male ancestors). The Y test could be more useful in figuring out Nicholas' ancestors or determining if your line is connected to Nicholas Ackley; the autosomal test would be more useful for general genealogy questions.
  • Decide which company to test with. The autosomal tests at both Ancestry and FTDNA cost $99 normally, so cost won't be the deciding factor. Both companies run specials from time to time; I've seen them both as low as $79. So the decision comes down to whether you want Ancestry's better genealogy tools or FTDNA's better genetic tools. If you've spent a lot of time and effort building your tree on Ancestry and want to take advantage of that, you'll probably want to go with them. If you'd rather have better tools to do analysis, you'll probably want FTDNA. If you want it all, you can test at Ancestry and either upload your results to GEDmatch for free and use their tools or FTDNA for a fee and use their tools.
  • For the Y test, your only choice is FTDNA. If you decide to do a Y test, you'll have the additional decision on how many markers to test; the more markers, the more expensive. The 37-marker test is $169, the 67-marker test is $268, and the 111-marker test is $359.
A few final words on next steps. First, you should know that I have no connection with either one of these companies, so I won't benefit whatsoever if you decide to test with either one of them. They both have a lot of my money!

Second, I know this can be a lot of money, so don't let me talk you into something you can't afford! I'm hoping that if you were already considering testing that this post will help you decide if you want to go through with it, and that if you do that you'll consider participating in some way with whatever project we can get going.

Finally, if you're already on FTDNA, join the Ackley project for free. It is easy to join and you might learn something. For those who want to participate in some way without dealing with FTDNA, I'd like to start an "informal" project on GEDmatch since they don't really have anything formal to conduct projects. That would entail just uploading your data to GEDmatch and then sharing your kit number with the rest of us so we can use their tools to do analysis. To get that rolling, I'll share my Ackley kit numbers here:

A235748 (me)
A590546 (my dad)
A833645 (my sister)
A913303 (my brother)
A588455 (my dad's Ackley cousin)

A couple of blog readers have already uploaded their results to GEDmatch and agreed to share. Their kit numbers are:

A527562 (John P. Ackley)
F364363 (Carol Harrison)

If anyone decides to upload to GEDmatch and wouldn't mind sharing your kit number, please post it in a comment on this post so we all can see it and use it for comparisons/analysis. I have created a separate "Ackley DNA Projects" page for the blog, and will keep an up-to-date list of all of the kit numbers there for easy reference. There is also a new page with instructions on how to download your data from Ancestry and upload to GEDmatch.

Additionally, I'm really hoping someone with the Hackley surname will test and share with us. That would really go a long way toward verifying or refuting the information in the post on Nicholas Ackley's ancestors.

Another thing to dream about is for people with Ackley ancestors in other parts of the world to test and add their results to the mix. If someone in England tested, that could give us a connection to the presumed origin of our family.

One final thought: I can't emphasize this enough -- using DNA testing successfully depends on having a well-developed family tree on both sides; i.e., both people in a match need to have done their genealogical homework. If two people match each other and one or both of them doesn't have a tree, they will learn next to nothing from the fact that they match. DNA doesn't tell you anything about who your ancestors were; it can only tell you if you share a common ancestor with someone you match. It is up to the two people who match to do the research to figure out who the common ancestor was.

Discussion Questions

  • Have you done a DNA test? If so, which company did you use? Which type of test did you do?
  • Do you have any DNA success stories you can share?
  • Would you be interested in participating in an Ackley DNA project?

Link of the Day

This link is to the Wiki for the International Society of Genetic Genealogy (ISOGG). There are over 500 articles about all aspects of DNA testing and its use in genealogy. Some of them get a little deep technically, but I have found most of them to be quite understandable and they have really helped me learn about genetic genealogy. I highly recommend it. By the way, membership in ISOGG is free (I just joined).

Quote of the Day

"It's the action, not the fruit of the action, that's important. You have to do the right thing. It may not be in your power, may not be in your time, that there'll be any fruit. But that doesn't mean you stop doing the right thing. You may never know what results come from your action. But if you do nothing, there will be no result."

--Mahatma Gandhi

Next Post Topic

Some "Other" Ackley Lines

No comments:

Post a Comment