Improved Gene Selection Method for Microarray Experiments
Author Information
Author(s): Yang James J, Yang Mark CK
Primary Institution: Henry Ford Health Sciences Center, Detroit, Michigan, USA; University of Florida, Gainesville, Florida, USA
Hypothesis
What is the best method to estimate the number of true null hypotheses in microarray experiments to control the false discovery rate?
Conclusion
The proposed method has better power to identify differentially expressed genes compared to existing methods while controlling the false discovery rate.
Supporting Evidence
- The proposed method showed higher power in simulations compared to existing methods.
- In real data analysis, the method identified more differentially expressed genes while maintaining the desired false discovery rate.
- The method effectively controls the false discovery rate in microarray experiments.
Takeaway
This study presents a new way to find important genes in experiments that look at many genes at once, helping scientists make better discoveries.
Methodology
The method uses a combination of test functions and permutation tests to estimate the number of differentially expressed genes.
Potential Biases
Potential biases may arise from the assumptions made about the distribution of p-values.
Limitations
The method may be conservative in certain scenarios and relies on accurate estimation of the number of true null hypotheses.
Participant Demographics
The study used data from prostate cancer patients, with 41 tumor samples and 41 matched normal samples.
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