Calibration Discrepancies in Affymetrix GeneChips
Author Information
Author(s): Harrison Andrew P, Johnston Caroline E, Orengo Christine A
Primary Institution: University of Essex
Hypothesis
Which calibration steps in Affymetrix GeneChips result in the biggest uncertainty in reported differentially expressed genes?
Conclusion
The choice of expression measure is the major source of calibration uncertainty in Affymetrix data analysis.
Supporting Evidence
- The study found that the choice of expression measure significantly affects the lists of differentially expressed genes.
- Using Z-scores provided a clearer diagnostic than fold change for identifying significant genes.
- Calibration protocols can produce different lists of significant genes, highlighting the importance of method selection.
Takeaway
When scientists analyze gene data, the way they combine information from multiple probes can change which genes they think are important.
Methodology
The study analyzed different calibration protocols for Affymetrix GeneChips, focusing on background correction, normalization, and expression measures.
Potential Biases
Potential biases arise from the choice of expression measures and the small number of biological replicates.
Limitations
The study may not account for all possible variations in calibration methods and their effects on gene expression results.
Participant Demographics
Data included samples from human, rat, and mouse.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website