I've come across the solution to a previously posted message about blast
output not being correctly parsed.
I think the reason is that in SRS5.1 the blast output is now expected to
be in the format that is derived from GCG's TOBLAST program rather than
NCBI's *2fasta and setdb programs.
The NCBI programs produce alignment headers that look like
>SW|acc|id des
whilst toblast generated headers like this
>SW:id acc des
where acc is the accession number, id is the entry id and des is the
description line.
So you either need to switch to using the GCG program to generate the
blast databases, or to modify blast.is to read the NCBI generated data.
Hope this helps.
Tim
--
Dr. Tim Dudgeon Phone: 01865 748747
British Biotech Pharmaceuticals Plc. FAX: 01865 717598
Watlington Road, Oxford, OX4 5LY, UK email: dudgeon at britbio.co.uk
-------------------==== Posted via Deja News ====-----------------------
http://www.dejanews.com/ Search, Read, Post to Usenet
