[Bioperl-l] SeqAnal.pm, Prediction parsers

Lorenz Pollak lorenz@ist.org
Wed, 6 Sep 2000 23:50:21 +0200 (CEST)

> I have to admit, I'm all for making a new "analysis" base class, but if we
> do this we should try to share it among the existing analysis classes, in
> particular BPLite and HMMER.

So if I understand that right, you want to put all modules that
parse reports and perform some analysis / prediction on that together
to a generic module and then a set of specialized modules an top of
>      Okay. I hope Steve won't mind that I'm going to steal a couple of
>      lines from SeqAnal.pm.
>      So, the issues are: (could the authors of BPLite [Lorentz?] and HMMER
>      (?) comment?)
>      1) What functionality (properties, methods) would you like to see
>      delegated to a generic base class, apart from program(), version(),
>      date()?

thats hard to say, because i don't know what all those different reports
(blast, hmmer) have in common that could go into a base class.
perhaps some properties about the sequence(s) reported there...

>      2) Do you already use a method parse()? There is then likely a name
>      clash.

there's no function called "parse". (and if so, it should be no problem to
rename that in BPlite, if there are some clashes)