[Biopython-dev] WIT and KEGG

Tarjei S Mikkelsen tarjei at genome.wi.mit.edu
Wed Aug 8 17:32:03 EDT 2001

> Yesterday, I downloaded your new code for enzymes.  I started 
> code, but WIT uses the KEGG format for enzymes.  So we may be 
> able to get by with one piece of code for both.

Yeah, I noticed that when I played around with WIT the other day.
I suspect that they're not only using the same format, but that
the enzyme record there are in fact the same as those in KEGG 
(or maybe it's the other way around, I don't know). - I haven't 
verified this though.

It makes sense to not duplicate the code, so we can either move
the shared parts into a module by itself, or you can just import
my KEGG code in your modules. 

> In your test files, what is the difference between the irregular and the
> sample files?.  

The KEGG distribution comes with a text file describing the 
record format. The .irregular files contains records distributed
by KEGG that does not conform to their description <sigh>.

> Did you manually strip out the HTML stuff?  

There was no HTML. You can download all the enzyme records in KEGG
in one big flatfile with no markup. If you want to pull down records
directly from a web page you can just strip the tags off in a simple
preprocessing step. There might even be a standard library call for

> I'll try to
> upload more test cases because sometimes I've seen bugs 
> on the tenth case.

That would be great.

> I hope its cooler where you live than in my area( 88 F ).

A little, it's about 80 at Logan now. They've warned us that we
might get up into the 90ies before the weekend though.
We'll just have to make sure the good old heat-shock proteins are
working :)


More information about the Biopython-dev mailing list