Abstract—Although microarray and deep sequencing technologies allow us to monitor gene expression at a genomic level, they also generate a relatively high false discovery rate, which limits our further exploration of disease mechanism. Furthermore, it is usually difficult to detect disease related genes with weak differential expression. It is well known that genes interact with each other during cellular and molecular processes. Therefore, disease related genes are not isolated. With large scale gene expression data, it is possible to evaluate the genome-wide co-expression information. This information is valuable to understand gene interactions. We have shown that differential expression analysis can be improved when this information can be incorporated. In this study, we demonstrate that genome-wide co-expression information can also clearly improve gene set enrichment analysis and classification analysis.