The pirtogcg failed for me too converting the nrl3d to GCG9 format, so
until
I get time to figure out what was wrong, I use a fasta version of the
sequence
data and then use GCG9's fastatogcg program to index it.
Okay, it has lost much of the sequence annotation, but then users could
always look for keywords in SRS5 anyway using nrl3d/pdbfinder/pdb or
whatever.
Regards,
Peter.
----------------------------------------------------------------------
Bioinformatics, mailto:p.woollard at hgmp.mrc.ac.uk
UK MRC Human Genome Mapping Project http://www.hgmp.mrc.ac.uk/
Resource Centre, Fax: ++44 (0)1223 494 512
Hinxton, Cambridge, CB10 1SB, UK Tel: ++44 (0)1223 494 523
----------------------------------------------------------------------
p.s. I have attached a no warranty pir2fasta perl5 conversion script.
Paul Roy wrote:
>> Hi GCGers:
> Does anyone else out there obtain the NRL_3D database (part of PIR)
> from the NCBI server? We did this last year (with GCG 8.1) and used
> DBINDEX to make .names .numbers and .offset files. An attempt to update
> using a current version of NRL_3D and GCG 9.0 resulted in DBINDEX giving a
> message like "no accession number in entry". The NRL_3D .ref files appear
> to have headers which are different from PIR1, PIR2, etc.
> BTW, the idea of including the NRL_3D database as part of PIR was to
> alert users of FASTA etc. to homologous proteins for which structural data
> exist. NRL_3D was not distributed by GCG nor included in pir.farm
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pir2fasta.pl
Type: application/x-perl
Size: 1688 bytes
Desc: not available
Url : http://iubio.bio.indiana.edu/bionet/mm/info-gcg/attachments/19970924/e5676b96/pir2fasta.bin
