Extracting Gene Expression Patterns from Microarray Data
Author Information
Author(s): Chou Jeff W, Zhou Tong, Kaufmann William K, Paules Richard S, Bushel Pierre R
Primary Institution: National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, USA
Hypothesis
Can a novel method for extracting gene expression patterns identify co-expressed genes that are biologically responsive to experimental conditions?
Conclusion
The EPIG method outperformed existing clustering algorithms in extracting biologically informative patterns and co-expressed genes from microarray data.
Supporting Evidence
- EPIG extracted 18 unique patterns and identified 2661 co-expressed genes from UV and IR-treated human fibroblasts.
- In a simulated dataset, EPIG accurately categorized profiles with 100% accuracy.
- EPIG revealed key biological categories related to p53-dependent cell cycle control.
Takeaway
The study developed a new method to find groups of genes that work together in response to different treatments, helping scientists understand how genes behave under stress.
Methodology
The study used a novel method called EPIG to analyze microarray data, comparing its performance against existing clustering algorithms like CLICK and CAST.
Limitations
The study did not address the performance of EPIG with datasets that have a very high number of clusters.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website