The Performance of Gene Expression Signature-Guided Drug-Disease Association in Different Categories of Drugs and Diseases.

Gene expression signature (GES) is a group of genes that shows a unique expression profile as a result of transcriptional machinery-related perturbations by drugs, genetic modification or diseases. The comparisons between GES profiles have been used to investigate the relationships between drugs, their targets and diseases with some successful cases reported. The rationale behind GES-guided drug-disease association is that if a medication can induce an opposite GES profile against that of a disease, it should possess the ability to reverse the gene expressions caused by the disease, and can be considered as a potential treatment of that disease. In this study, we data-mined the crowd extracted expression of differential signatures (CREEDS) database to evaluate the similarity of GES profiles between FDA approved drugs and their indicated diseases. The aim of our study is to explore the application domains of GES-guided drug-disease associations, that is, through the analysis of the similarity of GES profiles on known pairs of drug-disease associations, we can identify subgroups of drugs/diseases that are suitable for GES-guided drug-disease association for repositioning drugs. Our results suggest that GES-guided drug-disease association method is better suited for only some subgroups or pathways of drugs/diseases, such as drugs and diseases associated with immune system, non-chemotherapy drugs or mTOR signaling pathway, which showed significant higher correlations between their GES profiles.