[BioC] background correction

Ben Bolstad bolstad@stat.berkeley.edu
05 Feb 2003 07:23:30 -0800

On Wed, 2003-02-05 at 02:49, Dennis Kostka wrote:
> hi,
> i've been trying to get an idea about the implementation of the mas 5 
> backgound adjustment in the affy package. where it says in the sadd
> a(x,y) = max( i(x,y)-b(x,y) , nf*n(x,y) ) where 
> i(x,y) = max(i(x,y),0.5)
> i've just been able to find the first line implemented. where did i miss the 
> second line? or is it not implemented becaus .5 is so small that it should 
> rarely take effect? 

Cell intensities that low are extremely rare in my experience. In fact I
am not sure I have ever come across one. You'd be able to see if you are
getting anywhere near that with your data by looking at the
min(pm(....)) and min(mm(....))   where .... is your AffyBatch.

> (if that's the case, which is my guess, why is it in the 
> sadd?)

Affymetrix reason for doing things is not always clear.