[Bioperl-l] Re: MicroarrayIO proposal

[Allen Day]

> It *would* be cool to create MAGE-perl adaptor or wrapper classes that could
> live within Bioperl. This would enable bioperlers to make use of the
> MAGE-perl code for handling MAGE-ML data. I haven't thought much about
> specifics here, but would be interested if others have ideas along these
> lines.
> 
> Allen's modules serve mostly to help parse various non-MAGE-ML file formats
> containing expression data, a worthy cause which is outside the scope of
> MAGEstk. All we'd really need to tie into the MAGE world would be some
> adaptors that could interconvert his objects with MAGE-perl objects.

I might like to work on this.

> For the record, here's how Affy array layout data is mapped to the
> feature/reporter MAGE system:
> 
>    Single spot (cell) on an array = Feature
>    Perfect match cell + its mismatch cell = Reporter
>    All probe pairs (PM+MM) in a probe set = CompositeSequence

According to the definition above:

	Probeset.pm 	=>	CompositeSequence
	Probe.pm	=>	Feature

I haven't modeled the "Reporter" layer.

I'll have a look at the MAGE-perl classes and see if I can borrow some 
method/class names.

-Allen