Darke County Genealogical Researchers > DNA

DNA and genealogy made somewhat simple

By Karl Seitz

DNA genealogy can be scary for those of us whose last classroom contact with biology was decades ago. But it is not really that hard. You don't have to know all the science behind DNA testing to use its results in your genealogy. You don't need to debate the finer points of the differing DNA markers used by the various testing companies or whether a test for a particular ethnic background really does what it claims.

What you do need to understand is that the DNA test results for one individual doesn't tell you anything about the ancestors of that person. It is only when that result is compared with the results from other people that you get useful genealogical information.

There are two kinds of comparisons. The one with the most direct genealogical application is between two or a relatively small group of individuals to see if they share a common ancestor (inherited the same DNA). The other involves fitting your test results into all the other results available to the testing company. It is the latter that might give you some clues about where your ancestors came from.

The most common -- and most accurate -- DNA tests are limited to one line of ancestors each. Men can have their Y-chromosome DNA tested. This chromosome, which is only found in men, is passed down from father to son with only occasional mutations of its DNA. Both men and women can have their mitochondrial DNA tested. However, mtDNA is passed down only from mothers to their children, with even fewer mutations than Y-DNA.

It is also possible to test nuclear DNA, which is received from both parents and is not a direct copy of either. In theory, nuclear DNA should give you some idea about the ethnic background of your ancestors. However, knowledge of which DNA marker results originated in which part of the world is still in its infancy. The margin of error for these types of tests is high. And some testing claims are controversial. The margins of error should come down in the future as scientists gather more DNA results. Even then, there will be limits on how many generations you can go back before the results are more chance than science. If you're going to use nuclear DNA testing for genealogy, you need to learn more science than this article is intended to provide.

Both Y-chromosome and mtDNA tests can provide supporting evidence that two or more people are descended from a common ancestor, although by themselves neither can prove that a specific person is that common ancestor. For that proof, you need some traditional documentation. Either of the tests can, however, disprove a specific blood relationship.

For both social and scientific reasons, Y-chromosome DNA is the more generally useful DNA for genealogy. However, mtDNA can be the only useful DNA in specific genealogical situations. MtDNA can also be used to identify maternal-only ancestral origins long before any possible documentation.

Scientifically, Y-chromosome DNA mutates at a faster pace than mtDNA. Mutations are what create distinctive DNA family lines. Mutations can occur at any time, and while some DNA markers mutate faster than others, the average time between mutations is measured in generations. With its faster pace of change, Y-chromosome DNA has created a greater number of distinct family lines than mtDNA. That makes it easier to rule out a close relationship, which is sometimes the most important value of DNA comparisons.

Socially, Y-chromosome DNA parallels the inheritance of surnames in our culture. If two men have the same surname (variant spellings count) and matching DNA, they probably have a common ancestor. If these same men do not have matching DNA, their surnames are either coincidence and they are not related or one or both family trees have a ''nonparental event'' somewhere in the past. Nonparental events include undocumented adoptions, adultery, artificial insemination, anything that means the man who has been identified by society as the father of a child is not the biological father.

Tracing female lines is more difficult because of the name change that normally occurs at each generation. This cultural obstacle limits mtDNA tracing. However, if you want to skip thousands of years, mtDNA can probably tell you where your maternal ancestors at that time lived. You just won't know their names.

The test results for Y-chromosome DNA and mtDNA are stated in different ways. For Y-chromosome DNA, you get numbers corresponding to specific places (markers) on the chromosome. The numbers are the number of times a short sequence of DNA base pairs repeats itself at the specified location. For mtDNA, you get a series of codes that combine a particular marker on the DNA with a letter showing how the mtDNA sample differs at that marker from the Cambridge Reference Sequence of mtDNA. Cambridge University is where the original research was done that led to the identification of mtDNA haplogroups.

Haplogroups are the major divisions of human DNA. While mtDNA and Y-chromosome DNA each has haplogroups that are identified by letters, the two haplogroupings are separate classifications that are not linked to each other. Haplogroups are too broad to say that everybody in a particular haplogroup is related within historical times. However, two individuals who are not in the same haplogroup are definitely not related within historical times along their male-only or female-only ancestral lines.

Haplogroups are associated with geographic areas. However, the associations are so old that in most cases they have limited value for genealogy.

Choosing a company to test your DNA involves factors that are beyond the scope of this article. However, if two or more of you are using DNA to confirm or test your relationship, you should use the same company. It makes comparisons easier since different companies use different combinations of markers. There is considerable overlap in their choices, but enough differences to increase uncertainties.

DNA testers offer different levels of testing, with increasing prices as the number of markers being tested increases. Discounts are available from some companies if you join an existing family project or agree to the terms for starting a new project.

Unfortunately, the cheapest tests often don't provide sufficient information to differentiate between family lines. This is particularly noticeable on the Y-chromosome test where 10 or 12 markers will produce far too many matches among people who are obviously not related within historical times. You need 23 or more markers to reach good conclusions. Some companies are now offering to test from 33 to 67 markers, which should allow even greater distinctions between and among related lines.

MtDNA tests are offered at two or three levels, depending on the company. A basic test of the HRV1 bases might be sufficient when added to traditional documentation, but adding the HRV2 and/or HRV3 bases will allow for greater distinctions and might overcome gaps in the traditional documentation. At least one company is offering to test the entire mt-genome, which is probably overkill for most of us.

DNA testing is not a magic bullet for genealogy. It is rare that your DNA test results would allow you to connect to a family tree you didn't know existed. Yes, you might get lucky and match somebody who has the knowledge to close the gap between the ancestors you know and the common ancestor you share with that matching person. But you still have to have a certain amount of traditional documentation to get to the point where the other person can make the connection.

Without that second person's knowledge, a shot-in-the-dark DNA test might point you toward a certain geographical area as the place where ancestors originated. However, even that is uncertain given the mobility of the human race and the fact that DNA can remain unchanged for several centuries.

DNA results are most useful when you have a specific theory to test. For example, do you and somebody else with the same surname trace your ancestors to the same place, perhaps to two men who lived there at about the same time but without documentation to say they are related. Y-chromosome testing of a living male-line descendant of each man could prove or disprove a relationship. In practice and if possible, it would be best to test more than one line of descent in each family to catch any nonparental events along the way. (I can imagine a similar situation involving descendants of two women, but it would be harder to trace the lines of female-only descent in our culture.)

Matching results won't tell you the exact relationship of the two men, but they will tell you to keep looking for it. Nonmatches, if there are enough test subjects to rule out nonparental events, can tell you that seeking evidence of a relationship is probably a dead end.

It is also possible to get a mixture of matches and nonmatches that leave you scratching your head. That's when you need to revisit your theory to look for holes.

Not all matches are perfect. Sometimes test results will differ on one or a few markers. Then you have to look at how many markers were tested and the magnitude of the differences.

For example, if two Y-chromosome results match on 24 of 25 markers and differ by one or two numbers (repeats) on the 25th, they can be considered a match. This is an even more certain match if the difference is small and on a marker that is considered faster moving (it mutates more often than average). Even a 23 out of 25 match can be an overall match, particularly if there is a third test result that matches both of these on 24 out of 25 with differing sets of 24. The more test results you have, the easier it is to see a matching pattern among results that differ on one or two markers.

Those who want to explore DNA and genealogy further or just want to see how it was used in a specific case can use the following links. Your browser should open a new window or tab.

For a deeper introduction to using Y-chromosome DNA for genealogical purposes, visit DNA 101.

The DNA Testing Adviser provides a more comprehensive look at using DNA testing in genealogy.

To see how Y-chromosome DNA was used to prove the German village where my Seitz ancestors lived see The search for the German origins of Johannes Seitz on my Seitz Site.

To see how Y-chromosome test results can be displayed, visit Sides/Seitz/Sites/Scites/Sitz DNA results. The results in green near the top are from my Seitz family with the possible exception of the descendant of Eberhardt Frederick Seitz (kit 41836), whose German ancestry is unknown to me.