DNA Heritage - Tutorial for Surname DNA tests

On it's own, a single haplotype may give you an indication of where your

Y-chromosome originated from. There have been several scientific studies which have attempted to analyse the locations of haplotypes and haplogroups (find out about haplogroups in the Masterclass).

For example, the Atlantic Modal Haplotype (AMH for short) is defined by just six markers and is the most common haplotype in Western Europe:

Atlantic Modal Haplotype (AMH)

DYS19	DYS388	DYS390	DYS391	DYS392	DYS393

14	12	24	11	13	13

You may also compare your haplotype with those in the YHRD

database. Many scientific studies have entered their results into this database which has concentrated on Eurasian samples but now also contains American and East Asian, Eskimo Aluet and Amerindian samples. It can be searched at www.yhrd.org. The YHDR database uses up to eleven markers, all of which are tested by DNA Heritage.

In addition, Ybase (at www.ybase.org) is a useful tool for the

researcher and allows anyone who has had their Y-chromosome DNA tested to add their results and their genealogical details. Finding your DNA relatives is now much easier.

Y-DNA testing is at its most powerful when comparing two or more people

and the results overlaid onto the existing genealogical records.

Here is a simple scenario where 3 cousins of the same surname have been tested (diagram below). Two of them share a great-grandfather, and all three share a great-great-great grandfather.

In our family tree, only the males of the tree are shown. The red X shows where a paternal line has died out.

Paternal family tree (only males are shown)


	At some point way back in time, a single mutation in the Y-chromosome occurred. This mutation has left all the yellow males with this same mutation and can be detected when their DNA is analysed. When we compare their haplotypes we can observe the following:

Haplotypes of the 3 cousins

	Y-DNA STR Markers
	19	385a	385b	388	389i	389ii	390	391	392	393	425	426	437	438	439	460	461	462	GATA A10	GATA C4/ DYS635	TAGA H4

Cousin 1	14	12	17	12	13	29	24	11	13	13	12	10	15	10	11	10	12	12	12	25	27
Cousin 2	14	12	17	12	13	29	24	11	13	13	12	10	15	10	11	10	12	12	12	25	27
Cousin 3	14	12	17	12	13	29	24	11	14	13	12	10	15	10	11	10	12	12	12	25	27

As you read across, you'll see most of the numbers are the same except for the marker highlighted in light green. Cousin 3 shows the mutation at DYS392.

The results follow the paper genealogy that Cousins 1 and 2, who match each other at all 21 markers, are closely related. Cousin 3 is related, but more distantly so.

(If the above is a little confusing at first, don't worry! We'll do our utmost to explain what your results mean to you)

Genetic genealogy can substantiate the known, paper genealogy and help prove that two or more individuals, with the same surname, are connected by a common ancestor.

Estimating when that common ancestor actually lived is left down to mathematics and statistics. Studies show that although a mutation at any particular marker is a random event, it is expected to change roughly once every 500 generations. It is like a ticking clock, although this DNA clock doesn't always chime right on time.

Since DNA Heritage uses 43 markers, we can expect to see a single mutation once every 12 generations (500/43 = 11.6) - however your paternal DNA may have changed more recently, or is about to do so in a couple of generations time.

This simply means that the further back the MRCA (most recent common ancestor), the more mutations are expected.

If we have an exact match on all 21 markers, the average time when the MRCA lived is only 8.3 generations ago.

If there is a single mismatch (mutation) then this time increases to 20.5 generations.

For 21 STR markers

Number of mismatches	Average time to the MRCA (in generations)	95% confidence interval

0	8.3	0.3 to 43.9
1	20.5	3.0 to 68.0
2	33.2	7.7 to 90.5

Because mutations occur randomly, the average time to the MRCA can only be an estimate of when the MRCA actually lived. It can be a long time before or a long time afterwards - the 95% confidence interval is given in the third column (i.e. 95% of all cases will appear within that range).

So how many mutations/mismatches are needed to exclude a relationship? Well, as we said before, more mutations equates to a more distant relationship. With 21 markers, 2 mutations between haplotypes is a very much borderline result, but 3 mutations would usually exclude the possibility of a relationship between two individuals.

Now that we've tackled some statistics and haplotypes, let's put it into the context of a family research study.

North American office: P.O. Box 160, Carmine, TX 78932-0160 USA tel/fax: Toll free 866-7-DNA-DNA

European office: 40 Preston Road, Weymouth, Dorset, DT3 6PZ, UK tel:+44 (0) 1305 834936 fax:+44 (0) 1305 835925