DyeTerminator removal for Capillary Sequencer (ABI3100) - Edge

Jarmo Korkko jkorkko at tulane.edu
Tue Jul 3 13:36:34 EST 2001


We have been using pre-packed Edge Biosystems DyeTerminator Removal 96-well
plates for 7-8 months with really good results (and probably go back to them
again). Unfortunately the cost is quite high and therefore we tried
"do-it-yourself" Millipore MultiScreen-HV Filter plates with Sephadex G-50
Superfine (~$15 vs $50). We are using the same well-working platform (ABI
3100) to analyze the samples in both cases.

We have not been able to get the Millipore system to work well consistently
with our ABI 3100 Sequencer. Some samples are fine, some leak dyeterminators
(and/or other reagents) really badly. We have tried all the suggested
options for rinsing and pre-wetting etc., have tried different
centrifugation times. Constantly our signal strenghts are really high (which
is probably good or is it?). We have diluted the samples in regards to
reaction mix and DNA template which seems to help a little bit but we still
end up running some of the samples again after re-diluting them.

Does anyone have other ideas?

Our situation is complicated by the fact that our samples come from a number
of people (core lab) and the sample difficulties seem to be lab specific.
That did not use to be the case with Edge plates (if sample was to work the
results were nice). Now we are experiencing premature loss of resolution,
broad labeled nucleotide(?) peaks in the beginning (first 20-30-bp), signal
strentghts are out of control (up to 5000!!! with 1:10 diluted reactions
[~300-bp PCR products) and unpredictable.

I don't know if the signal is coming from dyeterminators that didn't get
removed or from other reagents that weren't removed in the process. Or do
the Edge columns really bind that much perfectly good product in their
columns? Is the loss of resolution from too much DNA trying to go through
the capillary or are salt etc remnants from the sample causing the trouble.
Edge plates have higher columns than Millipore, that might play a role..

We have decent results with our DNA robot when samples are quite diluted and
homogeneous, still consistenly some peaks in the first 20-30-bp.

Any comments are greatly appreciated.



Jarmo Korkko, M.D., Ph.D.
Research Assistant Professor
Tulane University Health Sciences Center
Center for Gene Therapy
1324 Tulane Avenue, SL99
New Orleans, LA 70112
Phone intl+1-504-988-7708
Cell Phone 504-388-3436
Fax 504-988-7704


More information about the Autoseq mailing list