🚀 go-pugleaf

From Geoff Hughes:

>Has anyone ported Mike Cherry's gb2ace.c program for VMS to a UNIX platform?
>It would be extraordinarily helpful to our efforts.

Hmm, well not exactly, but I've done some stuff that might be useful.
I've been writing conversion code in Perl.  This language was basicly 
designed to do just this kind of text manipulation/conversion, and I 
much prefer it over C for dealing with text.

At the moment the only thing I've written that processes genbank info 
is part of a collection of short Perl programs called by a shell script.
They are meant to periodicly search genbank for entries that match 
keywords (kept in a separate file), extract just the name and the sequence 
information, and append this info to a local blast-searchable database.  
For the Drosophila genome project, the keywords are various vectors, 
transposable elements, and repetitive stretches of DNA.  Other programs 
in our data-flow then search this database any time new sequence 
information comes in, in order to filter out sequences that match these 
(presumably undesirable) sequences.  However, this collection of programs 
is still rather rickety, so at the moment "periodicly" means when I get 
the chance to sit down and monitor its progress.  Mainly this is 
necessary because the keywords are too inclusive, so once I write up a 
better interpreter of the keyword file (one that can recognize that some 
entries indicate _particular_ sequences to extract, rather than 
general keywords) we should be able to have the script run automaticly
once a week or so to keep the filter database up to date.

I've also written some perl scripts that extract basic info from blast 
output and put it in .ace format, and a few other similar conversions.
Although Perl syntax can be odd, it is very powerful for this sort of 
thing.

						Gregg Helt