DNA ANALYSIS: The Newest Genealogical Tool?
 
 by John Thompson Johndnaguy@aol.com
 
The good news is that your entire family history is stored 
securely and unambiguously in your DNA. The bad news? It is not 
so easy to decipher and it covers only the identities of your 
ancestors, omitting details on their lives. So what can 
genealogists expect from DNA in the near future? To understand 
this, a little knowledge of genetics is required (but not much).
Any pair of unrelated people has exactly the same DNA sequence 
at about 99.8% of the three billion positions in human DNA. 
Related individuals have an even higher level of identity. Close
family relations can be confirmed by comparing the DNA sequence 
at positions that tend to vary more often than average (the 
differences can be either in the sequence of DNA present or its 
length). It is relatively easy to determine identity (like in 
forensics) or very close relationships (like determining 
paternity) but it becomes progressively more difficult to 
analyze more distant relationships.
 
Each person has 23 pairs of chromosomes in nearly all the cells
in their body with one of each pair contributed by the mother 
and one by the father. 22 of the pairs are virtually identical. 
One of the pairs is the sex chromosome that can be either XX 
(women) or XY (men). Each person gets one X from his or her 
mother and either an X or a Y from his or her father. In 
addition, most cells have hundreds of copies of mitochondrial 
DNA that is much shorter than the chromosomal DNA and comes only
from your mother. If the DNA behaved itself and tracked nicely 
from one generation to the next, molecular genealogy would be 
easy. However, there are changes in the DNA in every generation
(if you think transcribing census records is tough, try getting
3,000,000,000 base pairs of DNA right every time). Also, the 
individual chromosomes in the pairs recombine with each other, 
mixing up the parents' contributions in each generation. Thus, 
each of your great-grandparents supplied one-eighth of your DNA 
but their contributions are scattered throughout all of your 
chromosomes and not so easy to track, especially since each of 
your great-grandparents was 99.8% identical to the others to 
begin with! 
 
There are a couple of special cases that are easier to deal 
with. The simplest DNA to look at is the mitochondrial DNA 
(from only the mother) and the Y specific DNA (from only the 
father). More has been done with mitochondrial DNA because it is
easier to work with. It is much shorter and there are many more 
copies of it, but it can only be used to trace maternal 
lineages. If you and the person of interest share the same 
mitochondrial DNA sequence, you must have the same maternal 
ancestor some generations back. Not enough has been done to know
how much identity is needed to prove common ancestry because, as
mentioned earlier, DNA sequence changes slowly as you traverse 
generations. A very detailed account of how this type of 
information was used to identify century-old bones is presented 
in THE ROMANOVS: THE FINAL CHAPTER, by Robert Massie.
 
Y-specific DNA can be used to trace paternal lineages. Since the
mother does not have a Y chromosome, the father's contribution 
remains "pure." One good example of using Y DNA was confirmation
of a relation between Thomas Jefferson and the offspring of 
Sally Hemings (see the scientific journal NATURE, Volume 396, 
p. 27). It is important to note that with both the Romanovs and 
Jefferson/Hemings examples, DNA alone was not used to prove a 
genealogical relationship. A lot of research went into 
documenting historical data and providing a specific hypothesis 
that was then put to the DNA test. It is likely that this will 
remain the predominant use of the technology for the foreseeable
future. The use of DNA to establish broader ethnic/national 
heritage relationships is more doable in the near future, but 
the quality of the connections is highly dependent on well 
documented populations that are needed as a reference. For more 
information about DNA, the human genome project, and related 
topics, visit these Web sites:
http://www.ornl.gov/hgmis/publicat/publications.html
http://www.ncbi.nlm.nih.gov/disease/