Figuring out the molecular mechanisms underlying breast cancer is essential for the diagnosis and treatment of this invasive disorder. Hence it is important to identify the most significant genes correlated with molecular events and to study their interactions in order to identify breast cancer mechanisms. Here we focus on the gene expression profiles, which we have detected in breast cancer. High-throughput genomic innovations such as microarray have helped us understand the complex dynamics of multisystem diseases such as diabetes and cancer. We performed an analysis using microarray datasets with Networkanalyst bioinformatics tool, based on a random effect model. We achieved pivotal differential expressed genes like ADAMTS5, SCARA5, IGSF10, C2orf40 that had the most down-regulation and also COL10A1, COL11A1, UHRF1 that they had most up-regulation in four-stage of breast cancer. We used CentiScape and AllegroMCODE plugins in CytoScape software in order to achieve better insight into and figure out hub genes in the protein-protein interactions network. Besides, we utilized DAVID online software to find involved biological pathways and Gene ontology, also used Expression2kinase software in order to find upstream regulatory transcription factors and kinases. In conclusion, we have found that the statistical network inference approach is useful in gene prioritization, is capable of contributing to practical network signature discovery, providing insights into the mechanisms relevant to the disease. Our study has also developed a new candidate genes, pathways, transcription factors, and kinases that may be candidates for diagnostic biomarkers and also for drug design after experimental examinations.