ChipInspector
Introduction
ChipInspector extracts significant information right from the expression level of single probes from microarrays. Sophisticated analysis techniques and state-of-the-art genomic knowledge dramatically increase the number of significant features while simultaneously reducing false positive rates by an order of magnitude. ChipInspector has been shown to deliver useful results even from several years old formalin-fixed paraffin-embedded prostate cancer specimens.
ChipInspector uses the world's largest database of alternative transcripts and promoters to achieve superior signal-noise ratios in microarray analysis. It is unique in removing statistical and gene calling errors right at single probe level. This technology provides the basis for unmatched accuracy in significance analysis of microarray data.
Features
- Uses single probe expression levels as input
- Reduces false positive rates by about an order of magnitude
- Eliminates normalisation/interpolation problems
- Increases the number of significant features
- Supports Affymetrix CEL files (other platforms possible)
- Assigns probes correctly to transcripts and genes
- Accounts for alternative transcripts
- Direct transcript view with probe mapping
- Applies state-of-the-art genomic knowledge
- Extracts new information by mining previous microarray experiments
- Results are directly usable as input for BiblioSphere PathwayEdition
Example Output
The following is an example experiment with very good results based on Affymetrix probe set analysis. A SAM analysis with a false positive rate of 0.2 % yields an experiment specific network with 14 genes linked to four pathways. But some key genes known to be involved remain unconnected to observed pathways.
Re-evaluation of the identical experiment with ChipInspector on single probe level allows a SAM analysis with a false positive rate of 0.002%. The identical network analysis as above now shows 35 significantly regulated genes joined and linked to 31 pathways. All known key genes are included in the network.
