Selecting Informative Genes for Discriminant Analysis Using Multigene Expression Profiles
Author Information
Author(s): Yan Xin, Zheng Tian
Primary Institution: Russell Investments, Tacoma, WA, USA
Hypothesis
Can multigene expression profiles improve the selection of informative genes for discriminant analysis in breast cancer classification?
Conclusion
Methods that consider gene-gene interactions have better classification power in gene expression analysis.
Supporting Evidence
- MPAS and sMPAS methods showed a ~20% improvement in classification performance over conventional methods.
- Genes selected by MPAS had better marginal performance than other methods evaluated.
- Methods considering gene interactions identified important genes that would be overlooked by individual-gene methods.
Takeaway
This study shows that looking at how genes work together can help us better understand and classify breast cancer.
Methodology
The study used multigene expression profiles and backward information-driven screening methods to select important gene features.
Limitations
The performance of the methods may depend on the number of states into which the expression values are discretized.
Participant Demographics
The study involved 78 breast cancer patients, with 44 classified as good prognosis and 34 as poor prognosis.
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