Expression profiles resulting from the exon array can be classified into one of six classes (representative gene profiles are shown). Classification of the top 100 genes generated using the standard workflow. Red graph: exon expression in NSCLC; blue graph: exon expression in normal adjacent tissue; green bars: Splicing index per exon (logarithmic scale). (a) A gene that is neither affected by differential gene expression nor by differential splicing (none of the top 100 genes falls into this category). (b) False-positive result of a differentially expressed gene caused by exons with very low expression (arrows) outside the linear detection range. (c) Differential splicing (arrow) in the absence of differential gene expression. (d) Differential splicing (arrows) overlaid by differential gene expression. (e) Exons showing a constantly low expression value (arrow) that presumably measure the background signal leading to false-positives. (f) Exons showing a constantly high expression value (arrow) that presumably measure multiple targets leading to false-positives.
Langer et al. BMC Genomics 2010 11:676 doi:10.1186/1471-2164-11-676