>>>>> "Cornelius" == Cornelius Krasel <krasel at wpxx02.toxi.uni-wuerzburg.de > writes:
Cornelius> Harry Mangalam <mangalam at home.com> wrote:
>> What would you like to see for plasmid drawing? (Or anything
>> else for that matter..)
[..]
Cornelius> 1) A map compiler. This program would take a textfile
Cornelius> in some kind of format (e.g. GCG plasmidmap format) and
Cornelius> generate a map out of it (preferable output format is
Cornelius> something object-oriented, e.g. postscript, xfig or
Cornelius> similar, since this allows manipulation of the
Cornelius> generated format with something else. Output in a pixel
Cornelius> kind of format can be generated out of postscript with
Cornelius> ghostscript without any problems.)
There is something a little like this which will generate a linear map
in 6 frames, although it's not quite what you describe. It's a Java
annotation tool developed at the Sanger Centre and freely distributed
(see www.sanger.ac.uk/Software/Artemis/index.shtml). We use it as our
main annotation tool. Any features can be added in EMBL feature table
format. The output of Fasta/Blast/Gap/RE searches are parsed into EMBL
with Perl scripts and overlaid. As yet you have to get printed output
by grabbing screen dumps, so like I said, not quite what you described.
Cornelius> 2) A graphical program (kinda xfig-like) which allows
Cornelius> construction of plasmids by mouse-clicking. This
Cornelius> program would produce either the textfile mentioned
Cornelius> under (1) or the object-oriented graphics output. It
Cornelius> should be able to read the textfile as well and display
Cornelius> it on-screen. I wonder if one could build something
Cornelius> like that with Tk.
I was planning on using Perl/Tk with the BioPerl modules and EMBL.pm
for parsing EMBL entries obtained by efetch. The features could be
rendered on a Tk canvas which gives PostScript dumps without having
to generate PostScript directly. The Bioperl restriction cutting module
isn't fully developed as yet, so I was considering parsing tacg output.
Cornelius> 3) It would be nice to have a program which could
Cornelius> convert sequence information directly into the text
Cornelius> format mentioned under (1). I think tacg might be up to
Cornelius> that task. It would be even nicer if the program could
Cornelius> find some commonly used genes (like ampR in a plasmid)
Cornelius> by itself, either by doing some kind of sequence
Cornelius> comparison or by parsing genebank headers.
If you use the Perl, EMBL.pm already does this. It's a nice idea. Also
the ability to include DNA for which no sequence data is available so
that plasmid diagrams could still be drawn when uncharacterised
fragments have been cloned would be important for some lab workers.
If I come up with anything useful I'd be happy to offer it to anyone
working on an open source plasmid tool (it's early days, though).
--
Keith James -- kdj at sanger.ac.uk -- http://www.sanger.ac.uk/Users/kdj
The Sanger Centre, Wellcome Trust Genome Campus, Hinxton, Cambs CB10 1SA
