HELP! Odd Gel Pattern!!

Brian Foley brianf at
Fri Jun 3 12:33:30 EST 1994

brunker at wrote:
:     Instead of getting descrete bands on our electrophoresed 
: agarose gel, we've
: been getting "smeared" patterns...a dark band 
: followed by a gradient of lighter
: stains trailing back to the well. (I hope I'm describing this clearly).

	Won't it be nice when we can post actual pictures of gels into
our mail messages, so that we can all see what is going on?  Sometimes
a picture is worth 1000 words.  Actually, it is quite possible to
send photos through e-mail right now, but it requires a scanner to
input the picture, and knowledgeable users on both ends of the e-mail
	Anyway, your problem could be any number of things:
1) Overloading the gel.  But this usually causes streaks of DNA leading
back to the well, not a "gradient" as you say.
2) Too many cycles, too much template DNA, or too much polymerase.
Any one of these conditions can cause you to get a range of
PCR products from the expected size on up to 20-30 times the
expected size.  To diagnose this, take samples out of the reaction
after 5, 10, 15 and 20 cycles.  Run each on a lane of a gel.  See
if your product appears and then gets concatamerized in later rounds.
3) Bad gel.  But then your size markers would run poorly too.

:     Does anyone know what causes this, and how we fix it? 
:     These are PCR products digested by Nco1, and should produce not more than
: three nice strips each, not counting the primer lines. The marker lane looks 
: fine. The pieces we're looking at (we hope) are ~30, ~190 and `160 pb's.

	Why are you using agarose?  If I was looking at bands of that
size, I'd be using acrylamide.  I know that some more expensive
brands of agarose work for DNA in that range, but I always have
much better luck with acrylamide gels.  Extremely sharp bands and
much higher sensitivity (you can detect about 1/5 th as much DNA).

:     Has this happened to anyone else? Are we really producing an increasingly 
: large range of DNA strands? 

	Quite possibly.

:     How could such a fate have befallen us?!!

	This is nothing!  You should see the problems I've had in the
last six years!!!

:     Thanks,
:     Marla Brunker
:     brunker at

*  Brian Foley               *     If we knew what we were doing   *
*  Molecular Genetics Dept.  *     it wouldn't be called research  *
*  University of Vermont     *                                     *

More information about the Methods mailing list