DNA fingerprinting

[tivol at tethys.ph.albany.edu]

Dear Brad & Kathy & ...,
	DNA fingerprinting looks at the fragments produced by various endonuc-
leases, and compares the lengths of the fragments from different genomes.
These differences are referred to as "restriction fragment length polymorph-
isms" or RFLP's.  If there is enough DNA, pcr is not needed, but it can be
used to amplify small amounts of DNA for subsequent fingerprinting.  The com-
parison process consists of running electrophoresis gels and noting the pos-
itions of the fragments--the distance of travel is related to the length, and
the shorter pieces travel further.  A set of standards is also run on the
same gel.  There are enough things that can go slightly wrong that implement-
ation of a computer to analyse the prints could be very tricky.
	The basis of the usefulness of the method is that the endonucleases cut
DNA at specific sequences (usually about six base-pairs long), and random mu-
tations occur with a certain frequency within a restriction site--as these
specific sequences are called--or to produce a site when 5 of the 6 were
right and the mutation changes the sixth to the correct base, for example.
	Southern blot (or western blot, etc.) *are* the same methods in essense
although the exact technique may vary in some details.
	It can be seen from the foregoing that problems can occur with DNA from
mixed sources--blood from victim and perpetrator, e.g.  Furthermore, each 
RFLP may be shared by a large fraction of the population.  Only the combined
probability from having many RFLP's allows the statements like "The chance
that this fingerprint matches a person at random is one in a million."  As
is carefully stated by the more careful workers in this field, DNA finger-
printing is best used to *eliminate* a person as a source of the DNA.
	I hope this helps.  BTW, Southern blots are named for a person, so
Southern should be capitalized.

					Yours,
					Bill Tivol