IUBio

Running v.8 under v.9 GCG

Steve Thompson stevet at wsu.edu
Tue Feb 10 12:41:13 EST 1998


Hello all -

In regards to running EGCG v. 8 in GCG v. 9 environments, it gets even
messier than Rick, Fracois, and Guy allude to.  Granted, one does have to
switch environments back and forth between GCG v. 8 and v. 9 to achieve the
desired end, and there are various tricks to make this easier on the user,
such as those the above three people mentioned in their postings; however,
it is more complicated than just the environment.  A big problem comes up
when people want to use GCG v. 9 msf files in EGCG v. 8 because of the new
tilde (~) place holder character that GCG has adopted.  These are not
recognized as valid sequence characters by either EGCG v. 8 or GCG v. 8. 
Therefore, they need to be removed from your input msf files and replaced by
the standard period (.) GCG gap character.  The files then must be
reformated to correct their check sum lines.  I posted the following note in
our VADMS Newsletter several months back to formalize the solution to this
common problem at our site:

EGCG IS NOT COMPATIBLE WITH GCG V9!  DO NOT TRY TO RUN ANY EGCG PROGRAM
WITHOUT FOLLOWING THESE INSTRUCTIONS FIRST:

Before running any EGCG program (e.g. PrettyBox, PrettyPlot, ToPHYLIP,
SigCleave, etc.) FIRST initialize the old version of GCG with the "gcg8" 
command and then initialize "egcg":

% gcg8  (or run whatever script your systems manager has set up for this)
% egcg

Next, if dealing with a GCG version 9.0 MSF (multiple sequence format) file,
change all of the tilde's (~'s) to dots (.'s); here's one simple way using
the UNIX translate command at the system prompt:

% tr \~ \. < infile.msf > outfile.msf

Then, Reformat the sequence alignment using the MSF option to update the
file's checksums:

% reformat -msf outfile.msf{*} finalfile.msf

Now all EGCG programs will faithfully work with the alignment.

I hope this makes sense to everybody and that it eases a few frustrations.

                                            Cheers - Steve

                                Steven M. Thompson
              Consultant in Molecular Genetics and Sequence Analysis
        Visualization, Analysis & Design in the Molecular Sciences (VADMS)
             Washington State University, Pullman, WA 99164-4660, USA
                  AT&Tnet:  (509) 335-3179  FAX:  (509) 335-9688
                    INTERnet:  thompson at ribozyme.vadms.wsu.edu




More information about the Info-gcg mailing list

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