Hybridization Interactions in Microarrays Can Cause False Correlations
Author Information
Author(s): Okoniewski Michał J, Miller Crispin J
Primary Institution: Paterson Institute For Cancer Research, University of Manchester
Hypothesis
How do hybridization interactions between probesets in microarrays affect gene expression analysis?
Conclusion
Multiple targeting in microarrays is common and can lead to false positives in gene expression analysis.
Supporting Evidence
- Approximately half of the probesets on the HGU133A array were found to interact through multiple targeting.
- More than 10% of families of interacting probesets contained members annotated to different genes.
- Probesets capable of hybridizing to multiple transcripts can significantly increase correlation in expression data.
Takeaway
Sometimes, probes in microarrays can stick to more than one target, which can make it look like genes are related when they aren't.
Methodology
The study analyzed Affymetrix arrays to investigate the effects of multiple targeting on expression signals using both real and simulated datasets.
Potential Biases
The presence of multiple targeting can lead to false positives and misinterpretation of gene relationships.
Limitations
The analysis may underestimate the prevalence of multiple targeting due to reliance on well-characterized sequences.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website