DNA testing for genealogy purposes
Alexander Ancestry DNA Project
Roger Alexander
Volunteer Coordinator
July 23, 2012
Contents
Introduction to DNA
Which testing company should I use?
Surname projects
How many markers do I test?
What is the cost of a test?
Ordering a test
The DNA test kit itself
What happens after I test?
Understanding your results; interpretation and analysis
Will I match someone?
How do I get the ancestry information of the people I match?
Privacy issues
Is there a public website where I can post my results?
Your “deep ancestry” and haplogroups
Summary
Introduction to DNA
DNA (deoxyribonucleic acid) is a nucleic acid that contains the genetic instructions used in the development and functioning of all known living organisms. The main role of DNA molecules is the long-term storage of information. DNA is often compared to a set of blueprints or a recipe, or a code, since it contains the instructions needed to construct other components of cells, such as proteins and molecules.
DNA is passed from parent to child with rare changes (called mutations) occurring between two generations. For genealogical testing only two portions of your DNA are tested: yDNAand mtDNA.
yDNA is passed from male to male to male, etc. Thus it is used for tracing your surname ancestry. And only males can take the yDNA test.
mtDNA is passed from mother to both son and daughter, but it is only passed on by the daughter to her children. So either male or female can test for his maternal-only line.
In the chart below, the path of yDNA is in blue and the path of mtDNA is in red.
Males receive both Y-DNA and mtDNA
Females receive only mtDNA
Using DNA for genealogy purposes cannot replace good, old documentation. Paper trails are still the most important item in proving relationships. All the tested yDNA markers may stay the same from father to son for many generations, and, then, one, two, or even three markers may mutate at a single passage from father to son.
Therefore, with only DNA data and no other information about their ancestors, we can give only the probability that two tested men of the same surname have a common ancestor some number of generations in the past.
yDNA testing may be more valuable in telling us we are not related to someone with the same surname because it is quite good at doing that.
DNA is simply a tool to confirm who and who not you are genetically related to. But it gives focus to your research of your ancestry.
Which testing company to use
In 2000 commercial DNA testing for genealogy purposes began when Family Tree DNA ( became the first company to mass market DNA tests for genealogy purposes. Since then FTDNA has tested over 280,000 participants. Over 25 companies have started marketing those tests, many of whom no longer exist.
When we became volunteer coordinators of the Alexander Ancestry DNA Project in 2007, the project was recommending Family Tree DNA as the best choice of testing companies. We decided we wanted to evaluate all testing companies at that point in time to make sure we were comfortable with continuing to recommend FTDNA. Our criteria were:
- Customer service
- Quality
- Reputation
- Size of data base
- Support of surname projects
- Cost
FTDNA, in our opinion, was superior in the first 5 criteria. In terms of cost we found one could buy cheaper tests. But we felt that FTDNA was very competitive in its pricing, plus the value of the other 5 factors made it our overwhelming first choice. (We have no financial interests whatsoever in which company is used.)
At its founding FTDNA decided (uniquely, as it turned out) to organize its efforts around surname projects. The FTDNA website says prominently: Family Tree DNA encourages customers' participation in lineage and geographical projects. By transferring routine data management tasks to project administrators we are able to offer a discounted testing rate to those who order through a project.
Surname Projects
A surname project is administered by volunteercoordinators. The project is independent of the testing company. It sets its own policies. The Alexander Project, for example, accepts members from other testing companies. The coordinators perform organizational functions such as gathering the surname lineages of its members and comparing the DNA results of its members and running the project’s website.
FTDNA has over 5,700 surname projects. For example, the Alexander Project has 200 members and accepts only members who are genetically Alexanders. At last count the Alexander Project was about 70th in size at FTDNA, and incidentally the surname “Alexander” is about the 98th most common surname in the U. S. Our purpose in our volunteering for this work is simply and sincerely to help others understand DNA testing and to help others learn more about their ancestry. The website of the Alexander Surname Project is
How many markers do I test?
A “marker” is a location on the chromosome that is an inheritable indicator of identity or ancestry, and is measured in the quantity of “alleles” at that marker, usually a 2-digit number but occasionally a 1-digit number. The number of markers you test will determine the cost,andit will determine the quantity of DNA evidence you have to compare to others. Using Family Tree DNA and yDNAas examples, they offer tests of 12, 25, 37 and 67 markers.
It is recommended for testing for genealogical purposes at least 37 markers be tested. The basic reason is the more evidence you have to work with, the better off you are.
But even more important are the relative mutation rates of the markers in each of the 4 groups tested. Markers 1-12 are the most stable so at that level you can expect a very large number of matches. Markers 13-25 are less stable and yield a smaller number of matches but still an unmanageable number. Markers 26-37 are by far the most volatile. Markers 38-67 noticeably MORE stable than markers 26-37.
As an example, at 12 markers I have 858 exact matches. At 25 markers I have 116 exact matches. But at 37 markers I have NO exact matches and only one 36 out of 37 match with a known 3rd cousin.
Testing 67 markers enables one to refine comparisons and differences that may not show up at 37 markers. This level of testing is particularly helpful when there are a large number of members of a “DNA family” in trying to figure out whose branch came from whom. As more people test, the 67 marker test will become the “gold standard”.
What is the cost of a test?
At Family Tree DNA yDNA tests cost (if ordered through a surname project) as follows:
12 markers = $99
25 markers = $124
37 markers = $149
67 markers = $239
These prices are based on the tester joining a DNA Project before the test is ordered. If you do not join a Project the test costs more. In this way, FTDNA encourages participation in surname projects, a benefit to the tester.
If you are wondering if FTDNA ever has “sale pricing” or “promotion, for the last two years they have had a promotion with a discounted price of $119 for a 37 marker test, along with like discounts for other tests. Those promotion periods have been in June/July and November/December. The likelihood of sale pricing in the future is difficult to predict.
If money is a critical factor in the decision about how many markers to test, what many do, since they are not sure before ordering if they might have a match, is order a 37 marker test at the beginning. Then if they do match someone they can later upgrade to 67 markers for about the same price difference. AND they do not have to submit another DNA sample because FTDNA keeps your sample on file.
Ordering a test
To order a DNA test kit from Family Tree DNA, just go to This is the page for the Alexander Surname Project which will give you the best price available at the time.
When the page opens, scroll to the number of markers you want to order; it will give you the price. Then enter your mailing address where you want the test kit sent. Notice that they allow for two email addresses; so you might share the results notice and future match notices with another if you wish.
The DNA testkit itself
The next working day, FTDNA will mail you the test kit; it takes no more than an hour to un-package, understand the directions, rub the cheek, repackage, and send it back. They even include the return postage in the kit price. And it does not require a trip to the post office; your mailman will accept it.
The test kit you receive from the testing company consists of atiny, plastic toothbrush-shaped device and a tiny vial. The instructions ask you to rub the inside of the cheek for 30 seconds, then place the brush tip in the vial. Then return the kit to the testing company.
Test kit with scraping device and vial
You are also asked to sign a waiver allowing FTDNA to make public your markers and your contact email address.
What happens after I test?
FTDNA will give you your own "personal page" with your own password, and they will post target completion dates so you can follow the test's progress through their system. The results take about 3-4 weeks after you return the test kit. You get notice by email of the results, and, because you order through the Alexander DNA Project, they also send notice of your results to the project’s coordinators.
To be specific, your results are posted in groups of markers called “panels”. The first panel posted is markers 1-12, then a few days later the panel for markers 13-25 are posted, then a few days after that the panel for markers 26-37 is posted and so on.
Once you get notice of your results, you will logonto your personal page you will see something like this:
The system displays the names and email addresses of people that you match.
On the above screen if you click on the icon at the right of another tester’s name you will see an estimate of how closely related you might be to that tester, for example:
In comparing 67 markers, the probability that Roger M. Alexander and JohnAlexander shared a common ancestor within the last...generation is
.82% / 2 generations is
4.64% / 3 generations is
12.15% / 4 generations is
22.54% / 5 generations is
34.38% / 6 generations is
46.32%
7 generations is
57.39% / 8 generations is
67.04% / 9 generations is
75.06% / 10 generations is
81.49% / 11 generations is
86.48% / 12 generations is
90.27%
13 generations is
93.09% / 14 generations is
95.15% / 15 generations is
96.62% / 16 generations is
97.67% / 17 generations is
98.41% / 18 generations is
98.92%
19 generations is
99.27% / 20 generations is
99.51% / 21 generations is
99.67% / 22 generations is
99.78% / 23 generations is
99.86% / 24 generations is
99.91%
How closely are these two people related?
The reason it is an estimateof a probability instead of a certain, exact number of generations is because the markers mutate from generation to generation at an unpredictable rate. The estimated probabilities are based on the mutation rate of each individual marker.
The reason the estimates stop at 24 generations ago is that it was about then that surnames came into use, and is thus the beginning of “genealogical time”, (since the beginning of the use of surnames about 1200-1400 AD).
After you initially receive your results and notice of immediate matches, as time goes on and as more people are tested, you will get automatic emails from FTDNA advising you of new matches, with an email address through which you can contact these newly-found DNA cousins.
The advantage of DNA testing is that you will be able to identify and correspond with Alexanders with similar DNA results. In this way you can team with other "family" members of whom you had no previous knowledge and you can share information and research.
Understanding your results; interpretation and analysis
On your personal page you can see all of your matches, and you will be compared to them in terms of “genetic distance”, as shown in the above chart. Genetic distance is the number of markers you DO NOT match when compared to another tester.
To explain DNA results it is better to explain by example. Let me use the most common test: yDNA test for 37 markers. A “marker” is a location on your yDNA. It is expressed in the number of alleles, usually a two-digit number. Your 37 yDNA markers might look like this:
In comparing results of people who “match” each other it is conventional to highlight the markers the people DO NOT match on. Mismatched markers are indicated by yellow highlights when compared to your results on the top row below. It shows you are a genetic distance of 1 from the middle row of results (he mismatches on 1 marker) and a genetic distance of 2 from the bottom row of results (he mismatches on 2 markers).
The top row of markers is yours compared to two others whom you match.
FTDNA uses its data base to determine the average mutation rate of EACH marker. Based on which markers you and the other two examples mismatch on, FTDNA estimates:
The probability that you and your matches shared a common ancestor within the last number of generationsNumber of generations / 4 / 8 / 12 / 16 / 20 / 24
Probability / Match A / 59% / 89% / 97% / 99% / 99% / 99%
Match B / 29% / 69% / 89% / 97% / 99% / 99%
The reason it is an estimateof a probability instead of a certain, exact number of generations is because the markers mutate from generation to generation at an unpredictable rate.
Your ENTIRE “DNA family” (those whom you closely match) might look like this example. The markers and mismatches are shown here:
Notice you now begin to see patterns, which can be helpful in trying to figure out different branches of the family tree.
Will you match someone?
There is no guarantee you will match someone when you first get your results. One way to try to predict that is to check the lineages of existing members to see if you have a common ancestor.
In the Alexander DNA Project out of 200 members about 160 match another, i.e. about 80%. But even if you don’t match someone else, you will still learn who you do NOT match. And that is a lot more than you knew before you tested.
Although you may not closely match anyone when you first get your results, there are still benefits:
1. Eliminating false ancestral paths may be as valuable as finding real ones.
2. You can search for others who share your ancestry and encourage them to test so as to “prove” both your ancestries.
3. As DNA tests continue to grow in popularity, you will continue to have a larger and larger pool of testers for comparison.
How do I get the ancestry information of the people I match?
The project coordinator collects from members the members’ understanding of his direct-line Alexander ancestry. This record is maintained by the coordinator in Microsoft Excel format and is available to all project members. Matches are grouped together for ease of analysis.
In addition, that direct-line ancestry information is published at the project website In that public record confidentiality is maintained by NOT listing anyone born after 1900. Information in addition to the direct Alexander line typically is exchanged between people who match.
Gen 8 / Gen 7 / Gen 6 / Gen 5 / Gen 4 / Gen 3 / Gen 2 / Member / Gen. Dist.James, Sr, b.~1725-1730 / William b. ~1759, OR Matthew, b. ~1757, / John M., b. ~1790 or James C., b. ~1785, OR John, b. ~1791 / John Priestley, b. ~1825, Henry Co., TN, d.1906 / Joseph Riley, b. 1854, Henry Co., TN, d. 1933 / Berry, b. 1886, Henry Co. TN, d. 1976 / Nolan / (6th cousin?) / control @ 67
James Sr., b. 1725-35, d abt 1815 Blount Co TN. / James Jr., b. abt 1755, Anson Co NC. D 1826 Roane Co., TN / Lawson, b SC 1791, d Roane Co. TN 1861, / William Lawson, b. Roane Co. TN 1828, d. Loudon Co. TN 1877, / Joseph Peeler b 1868 Roane Co TN, d 1954 Louden Co TN / Ralph / (5th cousin) / 37/37
James Sr., b. 1722, d. abt 1808 Blount Co. TN / James Jr., b. 1755, Spartanburg Co., SC; d 1826 Roane Co. TN, / Thomas, b. 1797, Spartanburg, SC, d. 1878 Johnson Co. IN. / John Henry, b. 1832, Johnson Co. IN; d. 1878 Johnson Co. IN. / Joseph Armstrong b 1866 Johnson Co. IN; d 1926 Chicago IL; / Jarvis Madison / Jarvis Jr. / (3rd cousin) / 36/37
James Sr., b. 1722, d. abt 1808 Blount Co. TN / James Jr., b. 1755, Spartanburg Co., SC; d 1826 Roane Co. TN, / Thomas, b. 1797, Spartanburg, SC, d. 1878 Johnson Co. IN. / John Henry, b. 1832, Johnson Co. IN; d. 1878 Johnson Co. IN. / Madison, b. 1859, Johnson Co. IN, d. 1930 Franklin IN. / Mervill / Norman / (me) Roger / 64/67
Thomas b. 13 Oct. 1795 Spartanburg SC, d 16 Apr 1855 Oconee Co., SC; / Isaac Newton b 1836 Oconee Co., SC: d 1924; / DetrickBeiman b 1871; d 1933 Talladega, AL; / Thomas / Thomas / (?th cousin) / 36/37
William Andrew b. 1818 (NC or TN d. 1867 Polk Co., TN, / Thomas Andrew b. 1857, Union Co, GA / Creed / (?th cousin) / 65/67
James Wardlaw b Abbeville SC; d 1897, MS. / John M. b. 1841 SC? d. 1898 MS / George Hamilton b. 1871 MS d. 1921 MS / George / (?th Cousin) / 35/37
James Wardlaw b Abbeville SC; d 1897, MS. / David H / James Pascal / Charles / (?th Cousin) / 35/37
An example of a listing of the direct Alexander ancestries of some members who match each other.
Privacy issues
We are all concerned today about confidentiality. Concern is especially prevalent in genealogical DNA testing because of DNA-related crime news and shows on television combined with a lack of understanding on the part of the general public about DNA testing.
The yDNA test only examines specific sites on the male "Y" chromosome. AyDNA test is not the same kind of testing used for [1] paternity testing (uses father/mother/child autosomalDNA); [2] criminal identification (uses specific autosomal DNA markers identified by the FBI); [3] genetic disease detection (uses autosomal DNA). A good explanation of the four types of DNA can be found at: