Identification and normalization of plate effects in cDNA microarray data

Henrik Bengtsson

Centre for Mathematical Sciences
Mathematical Statistics
Lund Institute of Technology,
Lund University,

ISSN 1403-9338
Introducing a new way of visualizing cDNA microarray data we have identified a new type of systematic variation, which we refer to as plate effects. We believe that plate effects are due to non-biological differences in the cDNA clones products spotted onto the microarray slides. By comparing the consistency of all replicates (both within and between slides) after performing 42 different normalization strategies we claim that the plate effects arenon-negligible and should be corrected for in addition to the already known intensity dependent effects. For comparison between normalization strategies we introduce a novel robust genewise variability measure, which we call measure of reproducibility. In our case, we also found that not doing background correction improves the reproducibility significantly.
Key words:
cDNA microarrays; systematic variation; artifacts; plate effects; normalization