elliston@biovax.rutgers.edu (05/17/88)
From: <elliston@biovax.rutgers.edu> To All DNA Sequencers: I am looking for everyones little "favorite" tips for sequencing. I am putting together a little manual on sequencing and thought that it would be nice to include all the "tricks of the trade" that people on the Nets thought were of use. One particular thing I am looking for, is a method for using formamide gels (rather than Urea). I have heard that these are good for resolving compressions, but cant find a recipe or electrophoresis conditions anywhere. If I get some good responses and/or there is interest, I will summarize the results to the net. Thanx. Keith ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Keith O. Elliston | Waksman Institute Bitnet: ELLISTON@BIOVAX | P.O. Box 759 Arpanet: ELLISTON@biovax.rutgers.edu | Rutgers, The State U. of NJ AT&T: (201)932-3801 | Piscataway, NJ 08855-0759 USA ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ------
GORDON.BRADSHAW@BIONET-20.ARPA (05/18/88)
From: Toby Bradshaw <GORDON.BRADSHAW@BIONET-20.ARPA> Dear Keith, Three things limit conventional dideoxy sequencing: the resolving power of polyacrylamide gels, the ability of the polymerase to read through difficult regions, and the resolution of gel compression artifacts. As anyone who sequences large projects will confirm, the majority of time is spent not running reactions and gels, but editing. Editing hassles are mostly due to compressions. Let me save you a lot of trouble messing with traditional remedies (formamide gels, tempered glass "hot" gels, etc.) and say that the best way I've found in the 25kb of sequence I've done over the last few years is to use Sequenase with dITP instead of dGTP. This combination is nothing short of amazing. I just finished the traI region of F, which is about 58% G+C. It contains the usual "symmetrical" compressions where there is a compression which is not resolved on either strand alone, but which can be sequenced by taking the good (uncompressed) part from each strand. This type of compression is an editing headache, but does not usually lead to errors by experienced sequencers because it is clear from the beginning that there is a problem. The diabolical compressions are the ones which either smash one band into two, and therefore do not alter band spacing in their vicinity (giving no warning that there is a compression) or which are "asymmetrical", not resolving on opposite strands (rare but painful). In my experience, all these compressions are resolved on EITHER strand using Sequenase/dITP, EVERY TIME. The only problem is that, while it is always possible to count the number of bases, occasionally there is an ambiguous base or two. This can be minimized or eliminated by running the extension reaction for no more than 2min at no more than room temp (4C is OK), and the termination reaction for 15-30sec at 37C. The enzyme wimps out during protracted incubations, and it is critical that it be in top form during the termination or you will get bands across all lanes at every "G" (which is an "I", rather disliked by the polymerase). Therefore, extension reaction time/temperature should be the minimum consistent with good labelling. Sometimes an ambiguity may still exist, but is nearly always at a "G" and therefore resolves on the opposite strand even using dITP, since the offending base is now a "C".In an extreme case the template might need to be sequenced with dGTP, but I haven't had this happen to me. The reduction in editing when dI sequencing is dramatic and most welcome. The DuPont automatic sequencer, which uses fluorescent dideoxy dNTPs, would eliminate the only common artifact in dI sequencing (band across all lanes at "G"), since only strands actually terminated by a dideoxy are seen. I can't wait to try one of those. No, I don't work for U.S. Biochemical, but I tried to get some stock! Unfortunately, they are privately held. I have strong opinions on sequencing strategy, too, but I'll save that diatribe for later. Toby Bradshaw -------