Alas, poor "magic"! I knew thee well.

Warren Gallin wgallin at
Tue Mar 15 16:38:32 EST 1994

In Article <2m4kstINNllm at>, krasel at
(Cornelius Krasel) wrote:
>Martin Kennedy (mkennedy at wrote:
>: One point worth mentioning is that I still do 35-S manual sequencing; one of
>: the people in our Lab just tried getting some sequencing of these types of
>: preps done with 2 different ABI fluorescent machines, with no luck.  It seems
>: additional clean up steps are necessary to get this to work.
>It seems that automatic sequencing on ABI machines with Taq polymerase is
>more sensitive to contamination than manual sequencing with Sequenase.
>A guy in our lab brought his DNA and primers to the sequence service
>(who use an ABI sequencer) and had absolutely clean lanes, no band
>whatsoever; then he used the same DNA (which was, BTW, Quiagen-purified)
>with a Sequenase kit from USB, and was lucky. The people who operate
>the ABI machine have told us that this occurs from time to time. Has
>anyone (except Martin, see above) had similar experiences?

   When we switched to the ABI sequencer we had exactly this problem.  It
turns out that there were two problems. 1) for the cycle sequencing you need
longer primers than for sequenase.  ABI sells T3 and T7 primers that are
about 22mers, rather than 17mers.  We made our own; the longer primers
helped.  2) We were also doing alkaline lysis preps, a la Maniatis; bad
news.  When we used CsCl purified or MagicMiniprep purified material, we got
consistently 450-500 bp of reliable sequence.  We're trying out some of the
home-made versions now to see if they will work for the automated
sequencing.  I have colleagues who are sequencing PCR products directly. 
They electroelute using and IBI apparatus and  get excellent results.  My
take home messages from this experience were a) you need primers optimized
for your sequencing technique, and b) the Dye Terminators and/or Taq cycle
sequencing need purer template than many miniprep techniques will give you.
Warren Gallin,
Department of Zoology, University of Alberta
wgallin at

More information about the Methods mailing list