Re: Cleaning Plates/Bizarre Sequencing Results

Robert Chadwick (chadwick-1@medctr.osu.edu)
Fri, 07 Aug 1998 08:06:01 -0400

We have three 377-XL sequencers and have never had any problems related to glass
plates and/or buildup on the plates. We routinely get 700 to 800 bp of sequence
per reaction and never have had a problem with gel extrusion or gel
fluorescence. I suspect that most (or all) of the problems that users are
reporting are actually due to the techniques people use to clean the plates and
especially these bizarre methods people have come up with to strip them.
Our method to prevent these problems is to never use soap to clean your plates.
After a run we merely remove the old gel and rinse the plates thoroughly with
tap water and then a final rinse with dH20. Soap/NaOH/HCl are much worse
contaminants than any thing you use to make or run in your gels since
soap/NaOH/HCl contain a lot of ions which will alter the conductivity of your
gels. Also, organic contaminants in soap or alcohol may fluoresce and interfere
with your sequence data.

________________________________________________________________________

Robert B. Chadwick
Director, Genotyping/Sequencing Unit
Division of Human Cancer Genetics
Ohio State University
420 West 12th Avenue
Columbus, OH 43210

phone (614) 688-4783
FAX (614) 688-4761
e-mail chadwick-1@medctr.osu.edu

________________________________________________________________________

Bernd Weisshaar wrote:
> Problem description: there is a complete loss of signals in the lower
> region of the gel, and a "smeer" of green signal after that. In the upper
> part of the gel, the signals resume. Abi custom support only suggested
> better cleaning of glas plates, but the problem does not correlate with
> glas plate treatment. In one case, only half of the gel was affected (the
> gel was loaded in 4 steps, all probes were processed in parallel). If the
> samples are reloaded on a "good" gel, all is fine. But we have used the
> same chemicals and stocks for all gels .... where is the logic?

>Bernd -
>This is a problem that has plagued many sequencing labs.
>At one point, I scanned old emails and web sites and counted
>over two dozen locations reporting similar problems - including
>lots of the big sites, and ABI themselves. You will find that
>ABI is remarkably *un*helpful when trying to fix this. They
>really don't know the cause, nor can they suggest a reliable
>cure. I am quite disgusted by their lack of support and effort.
>The symptoms you report - blank areas of a gel and smeared
>data - are typical observations. You may also see (i) the blank
>region of the gel, followed by normal sequence, (ii) the smear
>region but no blank region, and (iii) any combination of these
>followed by a gradual restoration of normal data. You also
>pointed out that these symptoms may arise in some areas of a gel
>yet not others. You may see them associated with standards as
>well as "real" samples.
>While it may not seem logical, the greatest contributing
>factor really does seem to be the glass plates. There is weak
>evidence that the samples contribute to the failure, and that
>the specific gel mix you use can help or hurt. However, we can
>reliably resolve the smear/fade problem by switching to new
>glass plates. Obviously this is a short-term solution, since we
>can't spend $500 on plates every time we see a smear.
>It appears that there is some chemical buildup on sequencing
>plates. ABI does not know what it is. Several suggestions have
>been put forth by sequencing labs to minimize the effect:
>- Long Ranger gels are reported to suffer less from these
>artifacts.
>- avoid tap water when washing your glass plates.
>- The hotter the wash water, the less buildup you get.
>- various detergents are reported to help. You might
>try 'MultiTerge', available from Baxter Scientific.
>None of these is a complete solution. We have implemented all
>of them, and have reduced the frequency of fades/smears quite
>considerably, but we still occasionally get them.
>I would be MOST appreciative to hear of any real solutions
>to this problem. I am tired of seeing large numbers of samples
>(and large amounts of money) blown away by these artifacts.
>
>Bob Lyons
>University of Michigan
>
>-----------------------------
>Robert Lyons, Ph.D.
>Director, DNA Sequencing Core
>University of Michigan
>-----------------------------