DNA and genealogy - Tutorial links and other helpful information

[Paul Sheats]

I hope to provide some simple tutorial information in this thread if you are new to the concept of genetic testing for use in genealogy.

_________________________________


Use of genetics to aid in genealogy research was started in the year 2000. By comparing ones DNA to others who have tested, you can prove or disprove whether you are related to them or not, going back thousands of years. It can determine the likelyhood of how many generations ago two people shared a common ancestor. The better the match, the better the likelyhood of a more recent common ancestor. There are two main types of genetic testing used in genealogy.

The Y-DNA test traces your father's father's father (and so on) back in time. This is really the main type of DNA testing used in surname projects, such as the Sheets DNA project. Since Y-DNA is passed down only from father to son, you must be a male with the surname in question to join that surname project. However, if you are female, you can get a brother or uncle or other male blood relative with that surname to test for you.

Then there is the mtDNA test, which traces your mother's mother's mother (and so on) back in time. I will not focus on this, since it is not used in surname projects, but more in geographical location projects.

The test itself is simple. When you place your order, they will mail the apparatus for taking two cheek swab samples. You rub them around inside your cheek for about a minute to ensure a good sample. Then you mail them back and within a couple months, you get your results back.

There are different number of markers you can order. A 12-marker Y-DNA test is the least expensive, but many times this is just not enough markers to determine whether you match someone or not. 25 or 37 markers is recommended, because you might match someone on 12 of 12 markers, but later find out when you upgrade to 37 markers, you are now off by 8 markers and the chances of common ancestor within a genealogical time frame (600 years) is pretty nill. Basically, whatever you can afford is better, but you can always upgrade later to more markers. If you upgrade, they use the same samples you sent in the first time, so you don't have to take another DNA test.

The prices are listed at the project website on the "Recruitment" page. Here is the link:
http://worldfamilies.net/surnames/s/sheets/rec.html
These are the d-i-s-c-o-u-n-t-e-d prices for the Sheets project (if there were not already a Sheets project set up, the prices would be slightly higher.)

If you have Sheets ancestors, but no living Sheets cousins that you know about, you could still participate in recruiting efforts or by donating towards the cost of testing of participants. We could offer further d-i-s-c-o-u-n-t-s to participants in a variety of ways. This is our project, so I'm open to any level of participation. The recruiting is the hard part, but I feel it will be worth it in the end. Genetic testing can verify existing paper trails, and give you direction of where to search next (when the paper trail ends or is non-existant, such as in my own Sheets line). But it is pretty useless unless you have other Sheets individuals to compare against. Some surname projects have hundreds of participants, but they didn't get there overnight. It took a couple/few years.

[Paul Sheats]

Here are some benefits of DNA testing for genealogy research.

Benefits of Surname DNA Testing
* Eliminate or confirm relationships.
* Focus research towards related families.
* Direct research into a geographical area.
* Direct research into a specific timeframe.
* Establish country or region of origin.
* Confirm variant surnames are same family.
* Learn your family's pre-surname migration.
* Strengthen weak paper trails.
* Avoid pursuing false connections.


Here are a couple of tutorial links which are very good. If you have a high-speed connection, they even have a video tutorial (2nd link below).

Basics of Genetic Genealogy:
http://www.ftdna.com/tutorial_A.html

Tutorial links:
http://www.ftdna.com/dna101.html

[Paul Sheats]

I had an email question relating to my example of a person matching someone else at 12 markers, but then not matching at 37 markers. For clarification, I am posting a real example we ran into on the Sheets project.

Basically, each man that tests gets a value for each marker that represents the number of repeats of the DNA pattern at that marker. Below are the 12-marker results for members S-3 and S-4 at the project:

13, 23, 14, 10, 11, 14, 12, 12, 12, 13, 13, 29 (member S-4 results)
13, 25, 14, 10, 11, 14, 12, 12, 12, 13, 13, 29 (member S-3 results)

When you look at these, all of marker values match except the second one. One is 23, and the other is 25. 11 of 12 markers is pretty good match, but we weren't really sure if these two were related, so member S-4 upgraded to the 25-marker test (member S-3 had already tested to 37 markers). When comparing at 25 markers, it turned out that they only matched on a total of 17 out of 25 markers (written 17/25, which is not a good match). Then we knew that they couldn't possibly have been related going back several thousands of years.

My point in explaining this, is that a 12-marker test can give you an inkling of who you might match in a project, but usually it takes more markers to be sure. That's why I say, order at least the 25 marker test, and ideally the 37-marker test.

I hope that helps clarify the need for more markers. In the next message, I will talk about haplogroups.