Discovering negative correlated gene sets from integrative gene expression data for cancer prognosis
Along with the emergence and development of translational biomedicine, more and more genetic information has been applied in clinical practice. In recent decade, the discovery of genetic biomarkers for cancer prognosis obtains increasing attentions and many methods have been developed. The "element" methods use one or two independent genes to judge the Boolean status of disease. The "set" methods use general genetic biomarkers to classify patients into different risks as a whole. And the advanced "sets" methods use a group of different gene sets as biomarkers. However, the existing methods always concern positive correlations among genes ignoring negative correlations. Whereas the negative regulation, negative feedback, and functional repression are actually the important clues in cancer expression profiles. Therefore, in this paper, we propose to mine negative correlated gene sets (NCGSs) from multiple datasets, and use them along with the pure positive correlated gene sets for prognosis classification. The exploring experimental results have shown the encouraging promotion of cancer prognosis accuracy with NCGSs.