oozing gels 1 problem many of the same solutions

David Cain Dxcain at comdisco.com
Tue Oct 12 12:16:52 EST 1999

Dear Sequencers, 

I know that the oozing gel problem comes up periodically and that every lab has a different way of solving it but they all have a common thread.

The original solution was mentioned to me by Roger Derbyshire of PE in the UK, noting that the gel that was oozing was displaced from elsewhere in the gel this indicated a lack of efficient binding between the glass and the gel matrix.  Since the electrophoresis parameters are close to the maximum in the 0.2mm gels small imperfections that did not effect the 373 start to have an effect, namely the breaking down of the attractive forces between the glass and the gel.  The glass has a micro charge built up on it from rubbing and running under water that acts to attract the gel to it.  The charge is desirable so grounding yourself will not help the situation, what you need to do is promote the charge.  I have seen people do this by a variety of means:

rubbing the plates more vigorously than usual
leaving the plates under a running tap for an extra coffee break
longer cycles on dish washers

They all get the charge back and the problem goes away.

The charge is most commonly lost by the following:

soaking the plates in KOH without fully neutralizing afterwards
build up of residues usually TBE and/or  Acrylamide

As for using water in the comb space during the plate check I used to do something similar with a Li-Cor.  I removed the comb and applied a small amount of water to the well while I focused the laser and set-up the run.  This was an idea from some colleagues at MWG Germany who found that if you allowed the gel to reach temperature without electrophoresis the effect was to reduce the primer front to a minimal.  The gel did not swell up but there was also no history of oozing gels on the system either.

I hope this is of some help and clarification.


David Cain
Comdisco Lab and Scientific UK
Tel: +44 (0)20 8400 8265
Mob: +44 (0)7901 515094


More information about the Autoseq mailing list

Send comments to us at biosci-help [At] net.bio.net