Using Generalized Procrustes Analysis for Normalizing Microarray Data
Author Information
Author(s): Xiong Huiling, Zhang Dapeng, Martyniuk Christopher J, Trudeau Vance L, Xia Xuhua
Primary Institution: Centre for Advanced Research in Environmental Genomics (CAREG), Department of Biology, University of Ottawa
Hypothesis
Can Generalized Procrustes Analysis (GPA) effectively normalize cDNA microarray data without the assumptions required by other methods?
Conclusion
The GPA method is an effective normalization approach for microarray data analysis, particularly for boutique arrays where most genes may be differentially expressed.
Supporting Evidence
- GPA normalization showed lower variability than other methods.
- GPA is free from assumptions about data distribution, making it versatile.
- GPA performed better in reducing systematic bias in microarray data.
Takeaway
The GPA method helps scientists make sense of complex data from gene experiments by adjusting the data so it can be compared more easily, especially when many genes are changing.
Methodology
The study compared GPA with six other normalization methods using real and simulated microarray data, assessing performance based on variability and distribution similarity.
Potential Biases
The assumptions of other normalization methods may introduce biases that GPA aims to avoid.
Limitations
The study primarily focused on specific types of microarray data and may not generalize to all experimental designs.
Statistical Information
P-Value
p<0.01
Statistical Significance
p<0.01
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website