[Bioperl-pipeline] Requirements for pipeline installation

[shawnh@worf.fugu-sg.org]

On Wed, 28 Aug 2002, Elia Stupka wrote:

> > I can recommend either using the Generic Feature Database, which we are
> > refining which currently does allow storing of blast results
> 
> but that will not give you a simple way to view the data
> 

hmm. We can try having the first analysis do loading of bioperl-db or ensembl
taking fasta as inputs. just need to plug in the script. Then do a input
create on that to load the inputs and jobs. I need to get 
back to GMOD ahh..


> > or writing a hacky method that stores into ensembl without the need
> > for a contig.
> 
> but that will mean you will not be able to have the sequences associated
> with the annotation. Also, many calls later rely on having the sequence
> available. Imagine you start off by repeatmasking, and then you want to
> blast the repeatmasked sequence, if the sequences are in ensembl and the
> repeats are in ensembl it's a simple method call to get the repeatmasked
> sequence, if not, you have to roll your own code. And the same for anythin
> else.

I agree. XML is fast to write and I just wanted to show quickly that you can read from fasta if you want.
 But of course caveats will arise  later. But then, for some reason Bio::DB::Fasta
doesn't pick up the sequences for some reason when run in a module as opposed to 
 a script. Some issue with "tie" and scope.. I will need to look at it some more.  

shawn