|
Resolution: standard / high Figure 1.
(a) Fluorescence image of a hybridized Affymetrix GeneChip Mouse Genome MG430 2.0
array (GEO GSE12545). The chip surface divides into a grid of nearly 106 fluorescing probe spots. The bright horizontal stripe matches with probes the 25meric
sequence of which starts with triple degenerated guanines ((GGG)1-motif). Triple runs of other nucleotides are not associated with bright stipes. The
position of the respective probes are shown at the rigth border of the figure. (b)
Boxplots of expression measures obtained from the intensity data shown in panel a
using various preprocessing methods (see text for assignments). The boxplots are computed
separately for all probe sets (45,100) and for probe sets with at least two probes
containing the (GGG)1-motif (836, i.e. 2% of the total number). 'log I' denotes the distributions of raw
intensity data. Note that essentially all methods except Plier and partly mas5 are
unable to correct expression values for the (GGG)1-bias. The respective distributions of probe sets containing (GGG)1-probes are systematically shifted towards larger expression values compared with
the distribution of all probe sets.
Fasold et al. BMC Bioinformatics 2010 11:207 doi:10.1186/1471-2105-11-207 |