In <Zhd*eKsJo at news.chiark.greenend.org.uk> Tim Cutts <timc at chiark.greenend.org.uk> writes:
> One can certainly script gap4; you can access all of the assembly
> functionality and so on using tcl. For example:
>> #!/bin/env stash
>> load_package gap
Just a note. I used to use #!/bin/env for this sort of thing, but it appears
that on Linux env is in /usr/bin instead. Grr. So using env to avoid knowing
the pathname of stash doesn't help. Fortunately you can exploit a difference
between bourne shell commenting and Tcl commenting (Tcl joins lines ending in
backslash first, and then treats multiple lines joined in such a manner as a
single comment):
#!/bin/sh
#\
exec stash -f "$0" ${@+"$@"} || exit 1
load_package gap
...
> You can probably do this sort of thing for NIP too, certainly, typing
> "load_package nip" at a stash prompt does not return an error, so
> the functionality is there. Whether it's documented anywhere, of
> course, is another matter, although you could probably look at the
> tcl/tk scripts which make up the NIP4 user interface to find out what
> many of the commands are.
Nip4 and Sip4 are written in the same way so they can indeed be
scripted. HOWEVER (and this is a big however) many of the functions are not
split up in the right place, so asking for the restriction enzymes will call a
single function which calculates the positions and then plots them. The upshot
of this is that you still cannot use a script to avoid X.
This sort of problem has been addressed in our latest inhouse copy of Sip4 and
Nip4, which have now been merged to produce a new program named "Spin"
(suggestions for better names on a postcard please ;-)). This was mainly done
to allow a clean separation between algorithms and displays so that we could
export the results of our own algorithms and import the results from
others. Eg we now have a (rather experimental) interface to Emboss so that we
can use the emboss tools to produce sip4 and nip4 style plots which interact
which sequences and each other in exactly the same way that internal nip4/sip4
plots do. I haven't got a date for the release of this code yet though...
James
--
James Bonfield (jkb at mrc-lmb.cam.ac.uk) Tel: 01223 402499 Fax: 01223 213556
Medical Research Council - Laboratory of Molecular Biology,
Hills Road, Cambridge, CB2 2QH, England.
Also see Staden Package WWW site at http://www.mrc-lmb.cam.ac.uk/pubseq/
