Gliomas are the most common brain tumours in adult population with rapid progression and poor prognosis. Survival among the patients diagnosed with the most aggressive histopathological subtype of gliomas, the glioblastoma, is a mere 12.6 months given the current standard of care. While glioblastomas mostly occur in people over 60, the lower-grade gliomas afflict themselves upon individuals in their third and fourth decades of life. Collectively, the gliomas are one of the major causes of cancer-related death in individuals under fortyin the UK. Over the past twenty years, little has changed in the standard of glioma treatment and the disease has remained incurable. This study focuses on identifying potential therapeutic targets in gliomasusing systems-level approaches and large-scale data integration.I used publicly available transcriptomic data to identify gene co-expression networks associated with the progression of IDH1-mutant 1p/19q euploid astrocytomas from grade II to grade III and high-lighted hub-genes of these networks, which could be targeted to modulate their biological function. I also studied the changes in co-expression patterns between grade II and grade III gliomas and identified a cluster of genes with differential co-expression in different disease states (module M2). By data integration and adaptation of reverse-engineering methods, I elucidated master regulators of the module M2. I then sought to counteract the regulatory activity by using drug-induced gene expression dataset to find compounds inducing gene expression in the opposite direction of the disease signature. I proposed resveratrol as a potentially disease modifying compound, which when administered to patients with a low-grade disease could potentially delay glioma progression.Finally, I appliedanensemble-learning algorithm on a large-scale loss-of-function viability screen in cancer cell-lines with different genetic backgrounds to identify gene dependencies associated with chromosomal copy-number losses common intheglioblastomas. I propose five novel target predictions to be validated in future experiments.