[BioC] Re: Advice on print-tip normalization

Gene Cutler gcutler at amgen.com
Thu Sep 23 18:27:22 CEST 2004


On Sep 22, 2004, at 6:11 AM, Reimers, Mark (NIH/NCI) wrote:

> Hello Gene,
> I don't have any advice but some related observations based on looking 
> at regional biases on spotted microarrays. In the slide data that have 
> come in, there seems often to be a bias toward red on the top and 
> bottom edges of the print-tip groups, and a bias toward green in the 
> middle of the print-tip blocks. No explanation occurs to me, but this 
> effect is apparent on most of the arrays. One of our collaborators 
> claims the effect disappears with a more effective washing treatment, 
> but hasn't sent slide images.
>

What I'm seeing are largely overlapping periodicities in both channels. 
  The intensities start low at the beginning of a block, peak in the 
middle, and go down again at the end.  This repeats in each of the 48 
blocks across the slide, getting more exaggerated across the array.  
This is accompanied by a greater scatter in the intensity ratios, with 
much more noise in the center of the grids, where both channel 
intensities are highest.






Gene Cutler
Research Investigator
Bioinformatics

Amgen SF
1120 Veterans Blvd.
South San Francisco, CA  94080
(650) 244-2489



More information about the Bioconductor mailing list