We've never experienced undue problems with PCR samples, assuming
our *clients* do the basics:
- ensure there's a single band (or, weak alternative, gel-elute
a single-band to remove bogus amplification products).
- do some basic checks to verify that the proper product was
amplified in the first place (prove BOTH primers were essential,
restrict the band at a diagnostic site if possible).
- quantify the DNA via gel, NOT by spec (unless you have an
unusually sensitive spec). We request 20 ng/ul conc.
- make sure the primers are completely removed from the rxn.
We run BD 10 ul reactions with 40 ng of PCR product and 1.5 pMol
of primer. When PCR reactions fail to sequence, I sometimes check
out the situation with the client - and invariably find they've
slipped on one of the above. Most often its that they tried to
quantitate the DNA on a spec. Personally, I think subcloning the
PCR products is far more hassle that doing the PCR experiment more
carefully in the first place.
Bob Lyons
University of Michigan
-----------------------------
Robert Lyons, Ph.D.
Director, DNA Sequencing Core
University of Michigan
-----------------------------