Search for content in message boards

Family Tree DNA vs 23 and Me?

Replies: 31

Re: Family Tree DNA vs 23 and Me?

Posted: 31 Oct 2010 10:11PM GMT
Classification: Query
PDnger44,

I noticed that no one ever responded to your request for information on DNA testing. While this may too late to be of use to you, I will tell you what I know and hope it is of some help.

There is a significant difference between the way that FTDNA and 23 and ME identify male haploids that affects what further use of their data you can get out of it. Understanding the difference requires that one go into the history of genetic genealogy testing and understand the scientific basis of this difference.

Originally, geneticists figured out how to determine major divisions or changes in male human DNA which occurred over thousands of years as the Y-DNA genome developed various mutations over those time spans. They began to classify these divisions, called haploid groups, by analyzing for specific changes in the Y-DNA called Single Nucleotide Polymorphism (SNP)mutations. What they were looking for with the SNP was when a single nucleotide (Adenine, Thymine, Guanine, or Cytosine) change in the Y-DNA genome sequence at a specific spot in the genome. They identified these mutations by an alpha-numeric scheme giving them such names as M91, M42 M235, P13 and V12 mutations, etc. Once such a mutation occurs, which are relatively rare events requiring tens of thousands of years to occur, it becomes a permanent change to the Y-DNA of that person and all his descendant population. The point in time that these SNP mutations occurs defines a timeline of human genetic evolution and increasing diversity of genetic composition. Any given person has many SNPs that together create a unique DNA pattern for that individual, defining them as to a specific haploid division.

The testing service, 23 and ME, defines your Y-DNA haploid group and sub-clads by analyzing directly for these rare SNP mutations, and they have to analyze a bunch of them in order to narrow down those you are positive for and those you are negative for which they need to know to define your haploid all the way to its deepest clad subdivision. As a result, it generally costs $299 to get the answer from 23 and ME about what your haploid is.

FTDNA has a different approach that can define your haploid for only $99. They test for a different type of mutation which is much more common and frequent. This type of mutation is called a Single Tandem Repeat (STR) mutation. What happens in this type of mutation is a segment of your Y-DNA, called an allele (or Marker) gets damaged and a new copy of that allele is created and sandwiched in next to the old one. The new one is the only one that is active. All the other copies are non-functional. The number of repeat copies of a specific allele becomes the analytical test value reported by FTDNA for that Marker. So if there are 11 repeat copies of a particular allele, the reported value for that Marker would be 11.

Over many years of testing FTDNA has empirically selected a specific set of 12 alleles (or markers) of the Y-DNA genome which they have been able to use to predict your haploid with a high degree of accuracy, without actually testing the SNP markers that 23 and ME use to define your haploid. FTDNA is able to do this prediction because they have the largest genetic genealogy database in the world and have verified enough of their STR predictions with actual SNP analyses of the same sample that they have developed a STR pattern correlation algorithm that is 99% accurate in predicting your basic haploid group and often they can predict down to several subclads of that haploid based on your STR pattern, especially if you have the 37 marker STR test done.

Of course, Family Tree DNA can also test for the specific SNP that define the haploid they predicted from your STR values, but it will you cost additional money to have that verification testing done. However, FTDNA will not need to test very many SNP mutation points because they already will have predicted what basid SNP mutaions you have and will only test for the deepest subclad markers to verify your full haploid profile. That means that you actually spend the same amount of money with FTDNA obtaining a 12 marker STR result plus a verified analysis of your haploid down to its deepest defined sub-clad as you would with 23 and ME to get just the full subclad verified haploid results and no STR values.

My analysis of the usefulness of the end results between these two companies is that FTDNA gives you more useful data for the same money. For instance, in our haploid group within our surname we can tell the difference using the 12 marker STR values between descendants of two branches of our English family whose immigrant fathers came to New England separately in the early 1600s. We have also done the deep clad haploid testing of SNPs and found that we cannot tell them apart based solely on SNP results. We are, in fact all identical with respect to SNP results. Only the STR values show any differences we can use to group ourselves into the different lineages of our family group.

So, in a general sense, the SNP values obtained from 23 and ME provide only anthropologically useful information, whereas that information plus the STR values provide by FTDNA are useful genetic genealogy family group information that allow you to make comparision to the STR values of other men of the same surname and start to define different lineages within a descendant family.

I hope this has been helpful. If I am late in getting it to you and you have already made your decisions and commitments, then I am sorry I did not see your post earlier in the year when you first sent it out.

Best regards,
Mike Wright
SubjectAuthorDate Posted
pdinger44 23 May 2010 1:51PM GMT 
MichaelWright... 1 Nov 2010 4:11AM GMT 
pdinger44 1 Nov 2010 12:07PM GMT 
TRWhitaker62 17 Nov 2010 9:02PM GMT 
John Andrew G... 19 Dec 2010 5:46PM GMT 
debbiecruwysk... 21 Dec 2010 12:44AM GMT 
John Andrew G... 21 Dec 2010 12:59AM GMT 
John Andrew G... 21 Dec 2010 1:27AM GMT 
John Andrew G... 22 Dec 2010 5:45AM GMT 
debbiecruwysk... 22 Dec 2010 12:15PM GMT 
per page

Find a board about a specific topic

  • Visit our other sites:

© 1997-2014 Ancestry.com | Corporate Information | New Privacy | Terms and Conditions