ABRFers,
Perhaps someone has a suggestion about how to handle some
problematic samples in our DNA sequencing lab.
We occasionally receive samples with highly skewed nucleotide
compositions. Two situations produce such samples: organisms
with bizarre base compositions; and a procedure used to detect
the presence (or absence) of 5-Me-C using dimethylsulfate (?)
degradation of unmethylated bases. Both of these situations
produce highly skewed nucleotide compositions.
ABI's Sequence Analysis software has a problem with such samples.
It assumes that all nucleotides are present at "roughly" similar
numbers, and tries to normalize the color intensities before base-
calling. If in fact there are very few G's or A's, the software
just amplifies the background chatter up to the level of the
"real" peaks. The sequence is trashed.
On a 377, there's a trick that allows one to circumvent this
problem. A "hidden" menu can be called up in the SABasecaller*
that allows the operator to override the assumed nucleotide
composition. It has some bugs (mis-labeled nucleotides, strange
numbers), but we have been able to fiddle with the settings and
get these skewed-composition samples to sequence quite well.
Here's the problem - there's no such "hidden" menu on the 3700.
According to ABI, there's no way to adjust the expected nucleo-
tide compositional balance. I would *greatly* prefer to run
these on the 3700, but see no alternative but to fire up one of
our old gel-based machines.
Anyone have ideas on how we might run these on a 3700?
Bob Lyons
University of Michigan
P.S. -
The hidden menu is obtained by: double-click BasecallerPPC, then
hold down the option key while pulling the 'Edit' menu to 'Settings'.
*** Make sure you restore the changes to default values before going
on to normal samples! Click 'Default' in the Settings menu. ***
This archive was generated by hypermail 2b29 : Tue May 01 2001 - 14:07:05 EDT