[Bioperl-pipeline] Plant-Biotech pipeline 3

[Joachim H. Strach]

Hi,

after reflecting your answers, I have further questions.
I always refer to the genome_annotation_pipeline because it has a appropiate complexity.

First of all have a look at http://www.cosmoss.org/bm/pics/genome_annotation_pipeline.gif . I visualized the pipeline as I understood it.
Would be helpfull to dynamically visualize the pipeline while creating it with the biopipe_designer, wouldn't it?

1. What ist the meaning of !XXX!?
Is e.g. !INPUT! in iohandler id="2" the output from Monger id="1"?
Does !OUTPUT! refer to the ouput of the current analysis?
What about !INPUTOBJ!, !UNTRANSFORMED_INPUTOBJ!, !ANALYSIS!, ...

2. What's the idea behind the iohandler mapping?

3. The data monger ...
- always setups data for the following analysis?
- although e.g. analysis id="6" has a different iohandler and in my opinion no apperent link to the monger.
- always returns jobs or does it return what the setup_xxx function return?

Back to my pipeline:
4. How would you say is it meant to realize the following:
- storing blast-resulst (SearchIO) in a mysql-BioSql database
- filtering these results
- using the filtered results as input to a new blast (SeqI).
Espacially which functions and packages could I use? I have trouble to
identify suitable functions for my tasks. What is already there, what do I have 
to write by myself?
I did not find a package or function which provides filtering and converting of Bioperl Objects.



Thanks again

Joachim

______________________________________________________________________________
Bestes Testergebnis: Stiftung Warentest Doppelsieg fur WEB.DE FreeMail
und WEB.DE Club. Nur fuer unsere Nutzer! http://f.web.de/?mc=021182