[Biojava-l] blast parser

Simon Brocklehurst simon.brocklehurst@CambridgeAntibody.com
Thu, 02 Mar 2000 15:22:52 +0000


Dear All,

Re: Blast parsing system

Sorry for the delay here - things are a kinda hectic at CAT just at the
moment.   So I haven't been able to progress things quite as fast as I'd
hoped.

I thought I'd post a quick message to keep everyone abreast of where we
are. We have been refactoring our systems into what is a pretty neat
generic blast-like output parsing system. We've taken on board thoughts
and ideas from Mathhew, Wayne etc. in terms of the SAX-layer in the
design.  We have also taken on board the requirement from guys from SB
on the bioxml project for representing groups of related HSPs (also a
useful concept for the DBA program).

We've also weighed up the pros and cons of generic vs specific systems
for parsing different programs. We decided to go for a generic solution.
That is, we have objects (both Java value objects, and XML-based
representations) that capture all aspects of information associated with
the following pieces of software:

o Blast (NCBI)
o Wu-Blast (Wash. U.)
o Hmmer (Wash. U.)
o DBA (Sanger)

Hopefully this will do for starters.  More detailed info (along the
lines I gave in a previous post) when we actually load the system into
the cvs.

S.
--
Simon M. Brocklehurst, Ph.D.
Head of Bioinformatics
Cambridge Antibody Technology
The Science Park, Melbourn, Cambridgeshire, UK
http://www.CambridgeAntibody.com/
mailto:simon.brocklehurst@CambridgeAntibody.com