Bioperl: re: fasta def line problem
Thu, 27 Aug 1998 15:08:15 -0400
If there were a fast Perl-based XML parser, I'd vote for XML syntax.
Right now there's an interface to a C XML parser, but it is dog slow.
Has anyone looked at BSML (biological sequence markup language)?
Ian Korf writes:
> I agree with Lincoln, we need a written file forma standard. In some
> ways I like the fasta format because it's easy to parse knowing that the
> descriptor is on a single line beginning with ">". If the code is perl-ish
> then life is even easier. For example:
> >name=foo ID=23456 type=nucleotide ...
> No problem to use substitution to make this eval-able. I also like the
> tag=value convention because it is similar to HTML. Of course, if you're
> going to do that, you might as well be able to use arrays and hashes to
> get more expressive.
> Or even, why not just serialize your object with Data::Dumper and use that
> as the storage format. True, it doesn't look so good, but you could always
> have a prettyprinting text (and graphic) tool. I've used this to make a
> Bio::GSC client-server. It is a trivial operation if you keep the stream
> on one line and use the print and <> operators (a bit more difficult if
> you allow newlines). Anyway, if raw perl is too ugly, we could always make
> it just a bit prettier, do a substitution, and then inflate with eval.
> =========== Bioperl Project Mailing List Message Footer =======
> Project URL: http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/
> For info about how to (un)subscribe, where messages are archived, etc:
Lincoln D. Stein Cold Spring Harbor Laboratory
email@example.com Cold Spring Harbor, NY
=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/
For info about how to (un)subscribe, where messages are archived, etc: