Mining Gene Expression Data with PCA
Author Information
Author(s): Joseph C. Roden, Brandon W. King, Diane Trout, Ali Mortazavi, Barbara J. Wold, Christopher E. Hart
Primary Institution: California Institute of Technology
Hypothesis
Can principal components analysis (PCA) effectively identify biologically relevant gene sets from microarray data?
Conclusion
The PCA-based method successfully identifies numerous sets of biologically relevant genes, particularly in complex datasets with diverse conditions.
Supporting Evidence
- The PCA method identified significant gene sets associated with various conditions.
- Statistical enrichment analysis revealed biologically coherent gene sets.
- The method is particularly useful for datasets with many diverse conditions.
Takeaway
This study shows a way to find important groups of genes by looking at how they behave across different conditions, using a method called PCA.
Methodology
The study used principal components analysis to identify extreme gene probes and significant conditions in gene expression datasets.
Limitations
The analysis may overlook minor principal components that could still hold biological significance.
Participant Demographics
Tissue samples from three diagnosis groups: normal glucose tolerance (n = 17), impaired glucose tolerance (n = 8), and Type 2 diabetes mellitus (n = 18).
Statistical Information
P-Value
0.001
Statistical Significance
p<0.01
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website