Evaluating Gene Set Analysis Methods for Microarray Data
Author Information
Author(s): Dinu Irina, Liu Qi, Potter John D., Adewale Adeniyi J., Jhangri Gian S., Mueller Thomas, Einecke Gunilla, Famulsky Konrad, Halloran Philip, Yasui Yutaka
Primary Institution: University of Alberta
Hypothesis
This paper aims to compare the biological performance of various gene-set analysis methods for identifying differentially expressed pathways in microarray data.
Conclusion
The study found that SAM-GS, Global Test, and ANCOVA Global methods outperformed GSEA and other methods in identifying differentially expressed pathways.
Supporting Evidence
- SAM-GS, Global Test, and ANCOVA Global methods showed statistically significant differential expression for the majority of pathways.
- Areas under the ROC curves for SAM-GS, Global Test, and ANCOVA Global were above 0.80, indicating good performance.
- Other methods like Tian et al., Tomfohr et al., and GSEA had significantly smaller areas under the ROC curves.
Takeaway
The researchers looked at different ways to analyze gene data to see which methods are best at finding important patterns in cancer research.
Methodology
The study used three real microarray datasets corresponding to phenotypes defined by mutations in specific cancer genes and compared the performance of six gene-set analysis methods.
Limitations
The study's findings may not be generalizable due to the specific datasets used and the unknown true expression profiles.
Participant Demographics
The study analyzed 60 human cancer cell lines from the NCI-60 dataset.
Statistical Information
P-Value
p<0.05
Confidence Interval
95% confidence intervals reported for areas under the ROC curves.
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website