"FERRAN,edgardo" <Edgardo.FERRAN at TLS1.elfsanofi.fr> (Tel
61.39.96.00\(poste582\)) wrote:
>Dr. Chyh-Chong Chuang has wrote
>>>> Dear netters:
>>>> Does some one has the experience to extend gcg program to work on
>> personal sequences? This is the same to 'create' a personal database,
>> did any one has done that?
>>>>You may 'create' a personal database with the GCG-command dataset. For
<snip>
>Edgardo Ferran
>Sanofi Recherche
>e-mail: edgardo.ferran at tls1.elfsanofi.fr
Dataset is a good answer, but here's another suggestion.
If you don't have too many sequences, or expect to add to your set of sequences on a
regular basis, I find it easier to keep a file of sequence names up to date. You can
create this with a directory command and use it like a database. To make a file of
sequence names (list file) under unix, simply pipe the output of an ls command to a
file, eg a list file of all the sequences in your directory ending in .seq
ls *.seq > seq.fil
and then, to use this with, say, fasta, where you would specify a datalibrary
(eg: genembl:*), type @seq.fil
(Under vms, (well *I* still have to use it!) the same thing is accomplished by
$ dir/col=1/out=seq.fil *.seq )
When you add new sequences to your set, you simply recreate your file of sequence
names. You can run dataset -append, but it takes longer.
Just my opinion.
Cheers
Mathew
--
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Dr Mathew Woodwark mathew.woodwark at bbsrc.ac.uk
Molecular Biology Software Support Tel: 01582 762271
BBSRC Computing Centre Fax: 01582 761710
West Common, Harpenden
Hertfordshire, AL5 2JE, UK
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
