Global Differential Gene Expression in Cancers and its Implications for Building Robust Diagnostic Classifiers

Selecting differentially expressed genes (DEGs) is one of the most important tasks in microarray applications. However, the sample sizes typically used in current cancer studies may only partially reflect the widely altered gene expressions in cancers. By analyzing three large cancer datasets, we show that, in each cancer, a wide range of functional modules are altered and have high disease classification abilities. The results also show that modules shared across diverse cancers cover a wide range of functions, suggesting hints about the common mechanisms of cancers. Therefore, instead of relying on a few consensus individual genes whose selection is hardly reproducible in current microarray experiments, we may use functional modules as functional signatures to build robust diagnostic classifiers.