The role of Akt pathway in CD8+ T cell differentiation

Antibodies targeting T-cell inhibitory pathways such as the PD1-PDL1 axis are showing promising results in hard to treat cancers. Unfortunately, not all patients respond, about a third and currently there is no reliable means of identifying those individuals most likely to respond. What is known is that variation in effector T-cell frequency in tumours contributes to differences in overall response rates.

The research project will focus on identifying the underlying molecular processes that contribute to T-cell infiltration in cancers of the gastrointestinal (GI) tract. To provide an unbiased global view of these mechanisms, unsupervised learning methods will be applied to analyse RNA-Seq data in The Cancer Genome Atlas (TCGA). The approach to be used is based upon weighted gene co-expression network analysis (WGCNA), originally developed by Peter Langfelder and Steve Horvath, and modified at University of Southampton, to give improved resolution of immune cell subpopulations. Initially, WGCNA will be applied to TCGA-GI RNA-Seq data and all immune-related clusters categorised according to their main immune cell component and/or underlying biological process. To study the relationships among 'immune' and tumour clusters, we will summarise the expression profile of each gene cluster by its first principal component, termed the module eigengene (ME), and use mutual-information based distance measures to create eigengene networks. These eigengene networks will be overlaid with clinical trait data to examine how oncogene activation, for instance, modifies anti-tumour immune responses. The ultimate goal will be to integrate bioinformatic knowledge gained from these analyses into an executable Boolean model and to identify possible molecular pathways for intervention.