Glioblastoma (GB) is the most aggressive primary brain tumor in adults with an overall survival rate of only 15 months. Efficace treatment of diseases like brain tumors is hindered by heterogeneity not only between but also within patients, where multiple cell states exist, and by plasticity, where cell states can transition into another, favored by genetic aberrations and environmental cues. We are interested in unravelling the regulatory programs underlying this heterogeneity and plasticity. Therefore, we will exploit and develop machine learning and deep learning approaches to integrate and model multi-omics data at single-cell level. We foresee to identify regulatory mechanisms and key regulators of phenotypic plasticity, which could lead to more effective, personalized in the longer term.