Module-based breast cancer classification

Authors:
Yuji Zhang;Jianhua Xuan;Robert Clarke;Habtom W. Ressom
Affiliations:
Division of Biomedical Statistics and Informatics, Department of Health Sciences Research, Mayo Clinic College of Medicine, Rochester, Minnesota, 55905, USA;Department of Electrical and Computer Engineering, Virginia Polytechnic Institute and State University, Arlington, Virginia, 22203, USA;Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, District of Columbia, 20057, USA;Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, District of Columbia, 20057, USA
Venue:
International Journal of Data Mining and Bioinformatics
Year:
2013

Citing 9
Cited 0

A training algorithm for optimal margin classifiers

COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
Ant algorithms for discrete optimization

Artificial Life
Is cross-validation valid for small-sample microarray classification?

Bioinformatics
Outcome signature genes in breast cancer: is there a unique set?

Bioinformatics
PDZBase: a protein--protein interaction database for PDZ-domains

Bioinformatics
The MIPS mammalian protein--protein interaction database

Bioinformatics
FANMOD: a tool for fast network motif detection

Bioinformatics
Detecting functional modules in the yeast protein--protein interaction network

Bioinformatics
Peak selection from MALDI-TOF mass spectra using ant colony optimization

Bioinformatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

The reliability and reproducibility of gene biomarkers for classification of cancer patients has been challenged due to measurement noise and biological heterogeneity among patients. In this paper, we propose a novel module-based feature selection framework, which integrates biological network information and gene expression data to identify biomarkers not as individual genes but as functional modules. Results from four breast cancer studies demonstrate that the identified module biomarkers 1 achieve higher classification accuracy in independent validation datasets; 2 are more reproducible than individual gene markers; 3 improve the biological interpretability of results; 4 are enriched in cancer 'disease drivers'.