back to top
 
 
     
Here's a slightly more complex scenario, but one that is very common  
 

for most family names.

Let's take the fictional case of the Castle surname: People with this surname can be found throughout the world. There are thousands of Castle's around, and it is suspected that there were actually several different founding ancestors for this surname.

Also, it is suspected that the surnames of Cassell and Cazzelle are both related somehow to each other and to the Castle surname. DNA testing is an ideal method of providing many of the answers to such genealogical puzzles.

Several Castle's are asked to participate from each of the known lines of the Castle name. A couple of persons each with the names Cassell and

 
 

Cazzelle are also recruited at the same time to see if there are any long-lost links.

Here were the separate family lines and the number of participants from each line:

  
     
     
  Castle:  
  - John Castle of Durham, England, b.1545 (4 cousins)
  - Benjamin Castle of Devon, England, d.1700 (2 cousins)
  - Edward Castle of Northumberland, England, b.1701 (2 cousins)
     
  Cassell:  
  - Christopher Cassell of Lanarkshire, Scotland, b.1753 (3 cousins)
     
  Cazzelle:  
  - Daniel Cazzelle of Normandy, France, d.1693 (3 cousins)
     
     
 

From each line, at least two participants are needed. 'Non-paternity events' (adoption, infidelity etc.) occurs at about 2-5% every generation, so having two or more participants per line is vital to ensure that the correct haplotype is obtained for that particular line.

Also, as a general rule, the further back the suspected ancestor, the more participants are required. This is why 4 distant cousins were needed for the very old John Castle 1545 line.

After submitting their sample kits back to the lab, here were the haplotype results:

  
     

Bookmark and Share
Glossary - genetics terms explained
Cold Spring Harbor Laboratory
When humans faced extinction
Image gallery - pictures you can use on your own websites
Y-DNA haplotype results for the Castle Surname Project
     Y-DNA STR Markers (DYS and GATA)
Sample No.
Cladogram
Code*
Surname - Ancestor
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
 YLZ4734
A
Castle-JC
13
10
19
12
16
29
22
11
11
14
12
10
15
10
11
10
11
12
12
25
27
 YAC3237
B
Castle-JC
13
10
19
12
16
29
22
11
11
14
12
10
15
10
11
10
11
12
12
25
27
 YFD4354
C
Castle-JC
13
11
19
13
16
29
22
12
11
14
12
10
15
11
11
10
11
12
12
24
26
 YYD9114
D
Castle-JC
13
10
19
12
16
29
22
11
11
14
12
10
15
10
11
10
11
12
12
25
27
 YDX6549
E
Castle-BC
14
11
18
12
16
30
22
13
11
14
12
11
15
10
11
10
11
13
13
25
28
 YGJ3437
F
Castle-BC
14
11
18
12
16
30
22
13
11
14
12
12
15
10
11
10
11
13
13
25
28
 YRC2773
G
Castle-EC
13
10
19
12
16
29
22
11
11
14
12
10
15
10
11
10
11
12
12
25
27
 YYJ1892
H
Castle-EC
13
10
19
12
16
29
22
11
11
15
12
10
15
10
11
10
11
12
12
25
27
 YZF7984
I
Cassell-CC
13
10
19
12
16
29
22
11
11
14
12
10
15
10
11
10
11
12
12
25
27
 YEF2948
J
Cassell-CC
13
10
19
12
16
29
22
11
11
14
12
10
15
10
11
10
11
12
12
25
27
 YJH2245
K
Cassell-CC
13
10
19
12
16
29
22
10
11
14
12
10
15
10
11
10
11
12
12
25
27
 YDX8347
L
Cazzelle-DC
16
10
19
12
16
31
25
11
11
14
12
10
15
9
11
10
11
12
14
25
27
 YVN4848
M
Cazzelle-DC
16
10
19
12
16
31
25
11
11
15
12
10
15
9
11
10
11
12
14
25
27
 YPR4456
N
Cazzelle-DC
16
10
19
12
16
31
25
11
11
15
12
10
15
9
11
10
11
12
14
25
27
*the cladogram code is used to aid the construction of a cladogram (see diagram below)
           
Ancestor Key:
   JC  John Castle of Durham, England, b.1545  CC    Christopher Cassell of Lanarkshire, Scotland, b.1753
   BC  Benjamin Castle of Devon, England, d.1700  DC    Daniel Cazzelle of Normandy, France, d.1693
   EC  Edward Castle of Northumberland, England, b.1701      
           

back to top
     
 

Always the first feature to focus on is to see that the cousins from each individual ancestral line match up with each other.

If we take John Castle's line, and look at A,B,C and D, we can observe that the haplotypes of A,B and D are all identical. However, C is vastly dissimilar, mismatching by a total difference of six.

So is C related to A,B and D?

The answer is no. The reason being is that it takes many generations for six mutations to occur - too many for all four 'cousins' to possibly be related through their Y-chromosomes.

However, A, B and D do give us the 'ancestral haplotype' - the haplotype that defines that ancestral line.

If we look at each of the other individual lines, we see that they all match up with their respective cousins.

To see how all of the participants connect up to each other, we use software to analyse the data and produce a network diagram (called a cladogram):

  
     
 
Cladogram for the Castle Surname Project
A:B:D:G:I and J is a large group involving the lines of John Castle, Edward Castle and Christopher Cassell. The participants K and H are both only one mutation away and so are still part of the cluster. C is an unconnected outlier The Daniel Cazzelle line participants (M:N and L) are clustered together separately. The Benjamin Castle line participants (E and F) are clustered together but are distinctly separate from the other Castle lines.
roll over clusters for more info

back to top
back to top
     
   
 
You may note that DYS385a and DYS385b have been excluded from the analysis and cladogram even though there are differences observed.

This is because they are both 'multi-copy' markers which can be problematic when comparing two (or more) individuals and may result in false conclusions - hence they are left out of the analysis.

This is the same for all 'multi-copy' markers and is in line with many scientific population studies.
 
   
     
  We can see three main clusters and an obviously unconnected outlier (C).

On the left, (A:B:D:G:I and J) is a large group involving the lines of John Castle, Edward Castle and Christopher Cassell. The participants K and H are both only one mutation away and so are still part of the cluster.

The Benjamin Castle line participants (E and F) are clustered together but are distinctly separate from the other Castle lines.

The Daniel Cazzelle line participants (M:N and L) are also clustered together separately.

There is sufficient distance between each of the three clusters for them not to be related to each other.

So what conclusions can we draw?

  
 
1. The descendants of John Castle are quite likely related Edward Castle.
     
  2. The Christopher Cassell line is also most likely related to the
    above lines, indicating a possible name change at some point in the past (Castle can sound like 'cass-elle' in accents found within Scotland and the north of England).
     
  3. Participant C does not share the same Y-chromosome as his
    cousins, showing a separate paternal lineage.
     
  4. The Benjamin Castle and Daniel Cazzelle lines both have
    separate and distinct ancestors.
 
     
  It is important to understand that there will always be some limitations with current DNA testing, and should only be used in conjunction with the existing genealogy. Therefore the conclusions given above are not absolute certainties, but the DNA evidence does give very strong support for such conclusions.  
     
   
 

This information will have answered many questions that could not have been answered otherwise, provided new avenues for research and is a valuable new resource for other Castle's, Cassell's and Cazzelle's around the world to compare their own haplotypes against.

It is also something of immense value to preserve and pass on down to the next generation.

That's it! You've made it to the end of the tutorial…Well done!!

Numerous genealogists have already found their results have broken through many of their own brick walls, saving months of research and expense.

Hopefully you can see how powerful DNA testing is and the potential for your own Surname Project.

  
   
     
 

If you want to learn more on genetics with a wider scope, try the Cold Spring Harbor Laboratory website. Its graphics are excellent! Whatever your level of understanding, you're sure to learn something new.

The BBC also has a great site which explains how DNA has helped historians solve the riddles of the past.

The Rootsweb Genealogy-DNA board and mailing list has some very knowledgeable contributors, most of whom have gone through the same learning experience and have run their own Surname Projects. I would recommend subscribing to the Digest by email, or you can browse through the board.

We also provide the Masterclass pages as an additional resource. They are designed for those who wish to delve deeper into the intricacies of the Y-chromosome and keep up with the rapid pace of genetic genealogy research. For a more technical view, there are articles specifically on the Y-chromosome to be found at the Nature website.

Compare the DNA Heritage Y-DNA test with other companies.

Order your own individual Y-DNA test or start a Surname Project.

  
 
 
 
DNA Heritage® © 2002-2009     email:info@dnaheritage.com
North American office: P.O. Box 1028, Richmond, TX 77406-1028 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