Recent advances of innovative technologies and valuable data bases for proteome analysis enable us to identify extremely miner but important cellular proteins that related to the pathogenesis, with high throughput and quantitative manner. We established the differential analysis method with the combination of proteomic (2D-DIGE and cICAT/iTRAQ) and transcriptomic (DNA array) techniques using same brain tumor samples that show no pathological differences but significant variations to the chemotherapies, and tried to extract and identify the specific cellular signal cascade related to the chemotherapy sensitivities. For the basic information of human brain proteins, proteome database of brain tissues/cells were constructed with their 2D images and linked to the other brain database. After all of the differential analysis between chemotherapy sensitive and insensitive tumors using above technologies, the significant proteins identified were assembled, integrated, and subjected to the cellular signal network analysis. In the case of anaplastic origodendroglioma/astrosytoma: AOG, a series of proteins that changes their expressions and post-translational modification statuses, in relation to the tumor stages or chemotherapy sensitivities, were quantitatively identified. As glioma specific proteins, 738 proteins were identified in total, and 201 were related to chemotherapy sensitivities in the group of anaplastic gliomas. Using these protein data sets, the activated cellular signals in the chemotherapy insensitive glioma were extracted by the network analysis. They include several unknown signal cascades in gliomas, such as signals related to the specific cell cycle, apoptosis, and cell adhesions. These newly established differential analysis and data mining methods will be useful not only for finding of diagnostics for gliomas, but also for those of other brain diseases.