Wretched dinucleotide repeats

Andy Law Andy.Law at bbsrc.ac.uk
Tue Jun 11 03:53:04 EST 1996

In article <wolffie-1006961059330001 at>,
wolffie at u.washington.edu (John Wolff) wrote:

 >  In article <4okhac$ljk at sol.sun.csd.unb.ca>, sgriffit at rpc.unb.ca (Steve
 >  Griffiths) wrote:
 >  >  We're using selected bovine microsattelites as tools in forensic 
 >  > analysis of moose and white tailed deer material to catch poachers.As 
 >  > with any dinucleotide repeat system the real allele holds court to an 
 >  > xtra band(s?) as a result of T addition by Taq ,as well as the lower 
 >  > molecular weight stutter bands that I'm lead to believe result from the 
 >  > collapse or shift of repeat units.I've been dickering around with T4 
 >  > polymerase which is supposed to nip the T overhangs off the amplified 
 >  > product, but ,so far,the enzyme and 5X buffer supplied by Gibco don't 
 >  > appear to have much effect. Has anyone had any success in reducing the 
 >  > appearance of these additions or shortened products? Is there any way of 
 >  > preventing the stutter effect?
 >  If you can find polymorphic tri- or tetra-nucleotide repeats, they're a
 >  lot easier to read.  I've done quite a few different human dinucs and
 >  noticed a lot of variability.  Some just look downright weird.  Sometimes
 >  with tri- or tetra-nuc repeats, you have to look out for PCR apparently
 >  amplifying the smallest allele preferentially -- the smallest band is
 >  always darker, and sometimes the larger allele fails to show.
 >  BUT:  the stutter bands are a nuisance, not an insuperable problem.  The
 >  hardest part is to tell a homozygote from a heterozygote whose alleles are
 >  just 2bp different:  in such a heterozygote, the 2nd largest band will be
 >  DARKER than the largest, while in a homozygote, the largest band should be
 >  darker.  You have to be REALLY careful.  I'm sure defense lawyers would
 >  have a great time with these things!!

There are two issues here.

1) Stutter bands.

These are the result of 'slippage' during the amplification process. They
are pretty difficult to eliminate and occasionally may be helpful in
identifying genuine alleles from dye-overspill or other artefact i.e. a
real allele will have a stutter band. If two alleles differ in size by the
same amount as the allele and its stutter band, again you will see a
characteristic pattern of peaks (which are difficult to represent on a
character basis) which will be a small stutter band from the smaller
allele, a  large peak which represents the smaller allele product plus the
larger allele stutter band and a larger allele. Once again, this pattern
is helpful in concluding that the interpretation of the gel is correct.

2) The addition of extra bases to the end of PCR products by Taq.

The degree to which this is a problem differs unpredictably from marker to
marker, and from allele to allele in our experience. We believe that the
best way to handle the problem is to minimise the time available for the
unwanted additions to occur i.e. get the PCR reaction off the machine and
crippled as soon as possible. At present, we try to get all our PCR
reactions done and ethanol precipitated or EDTA treated within 20 minutes
of completion.

If you are doing forensic work, you will presumably have a detailed SOP
for the PCR reactions - I strongly suggest that the post processing be
included in the SOP and completed immediately the reaction finishes; if
that doesn't work, then try a different marker (or pair of primers).

Andy Law
( Andy.Law at bbsrc.ac.uk )
( Big Nose in Edinburgh )

More information about the Methods mailing list