Do you have to sequence both strands?

Brian Foley brianf at
Tue Apr 5 10:59:20 EST 1994

 In article <gosink-020494174955 at>, gosink at (John Gosink) writes:
: > Hi,
: > 						I was having a discussion with a person down the hall.  They claim
: > that there is _no_ regulation that says you have to sequence both strands
: > of a length of DNA (they are working with cloned PCR products) for
: > submission to Genbank and/or a referreed paper.  Do you have any references
: > on this subject?
: > 
: > 			                              -John
: > 
: > P.S.  They read a single strand, but make it a point to read each gel on
: > two different occasions and/or by two different people.
: -- 

	I have seen some of my sequences that look perfect from one primer
and also look perfect from the other primer that reads the other strand,
but the two do not agree.  That is to say, that both gels are perfectly
readable, with no indication of a compression or anything.  Yet one
of the strands DOES contain a compression.  If I use dITP with both
primers, the sequences are less readable overall (many places with bands 
in all four lanes), but it does show me which of the two primers gives
me the correct sequence.
	So, just because you can read a sequence without any ambiguity,
does not mean that the sequence is CORRECT.  One of my compressions
occurs only 34 bases from the 3' end of my primer.  It runs near the
bottom of the gel where it is very readable.  I could not beleive it
when I first sequenced the other strand and got a conflicting sequence
there.  At first I thought I must have sequenced a different plasmid
prep in which the E. coli had induced a mutation.  I have since proved
to myself that it is just a compression.

	So yes you must sequence both strands if you want to be
sure you have the correct sequence.  If you want to just crank out
some data and don't care if it is right or not, you can get away
with just doing one strand.  Sequencing just one strand
might be OK if you allready have the sequence and are just sequencing
a few clones to look for a mutation or something.  But I would
not recommend publishing such work or submitting it to GenBank.
*  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