Dysregulation of the PI3K/AKT/mTORC1 signalling pathway occurs in up to one-third of all sporadic human cancers. Paradoxically, oncogenic activation of this network in non-transformed cells promotes senescence, which poses a significant barrier to malignant progression. Understanding the basis of oncogene-induced senescence will provide insight into cancer development and potentially identify novel therapeutic targets. Our previous work showed that hyperactivation of AKT results in senescence via a p53 and mTORC1-dependent mechanism (Astle et al, Oncogene 2011). To determine the genetic changes required to overcome AKT-induced senescence, we performed a genome-wide RNAi gain-of-function screen using multi-parametric readouts including cell number, proliferation, and senescence-associated beta-galactosidase (SA-βGal) staining. In addition, we have utilised machine learning and automated image analysis methods to classify distinct phenotypes of cells with SA-βGal staining. Using various genetic approaches, we are investigating the role of candidate tumour suppressor gene targets and how they can be exploited in cancer.