Thermal cyclers for automated sequencing?

John H McDonald mcdonald at strauss.udel.edu
Wed May 25 14:11:23 EST 1994


Yes, you can use an Idaho Technologies thermal cycler (the ones that use
glass capillary tubes) for sequencing  for an ABI 373 automated sequencer.
The conditions you use will depend on whether you're doing dye-primer or
dye terminator sequencing.  
	If you're doing dye-primer sequencing, there was a paper in a recent
issue of Biotechniques describing the conditions (Swerdlow et al. 1993,
Biotechniques 15:512-519).  The modifications they made were to silanize
the capillary tubes; to use 15 cycles of 95 C for 0 seconds, 50 C for 0
seconds, 70 C for 20 seconds followed by 15 cycles of 95 C for 0 seconds,
70 C for 20 seconds; 10 ul reactions; and 50 ug/ml BSA in the reaction. 
They got good looking sequence, and said that the sequence in an area of
compression was better on the Idaho than on a slow old Perkin Elmer.
	We've been doing dye-terminator sequencing.  Unlike Swerdlow et
al., we don't silanize the capillary tubes, and we add 500 ug/ml BSA
(final concentration) to the reaction, not 50 ug/ml, to prevent the Taq
from binding to the glass.  Like them, we use 10 ul reactions, not 20 as
recommended by ABI.  This cuts the cost of the expensive reagents in half,
of course, and still gives plenty of signal.  We use 25 cycles of 94C for
0 seconds, 50C for 0 seconds, and 60C for 3 minutes.  The long extension
time is necessary for dye-terminator sequencing because it uses
thiol-modified dNTPs, and Taq processes these more slowly than regular
dNTPs.  We haven't done experiments to see whether we can cut the
extension time shorter than 3 minutes.  We use the phenol-chloroform
method to remove the excess dye-terminators, so the BSA in the reaction is
also removed.  I don't know if it would cause problems if you clean up
with columns, but if so you could probably use considerably less BSA than
we do. 
	As usual, no financial connection with any company, etc. etc.

John H. McDonald
Department of Biology
University of Delaware




More information about the Methods mailing list