Genome-wide regulatory networks of T cell differentiation
Speaker: Mika Gustafsson, Associate professor in bioinformatics, IFM, Linköping University, and Centre for Individualised Medicine
The cost for genome-wide measurements in molecular biology (‘omics) has drastically decreased during the last 10 years. Systems medicine analysing the measurements from thousands of molecules will therefore likely become routine for medical doctors within the clinics. Potentially, this could lead to preventative and personalized medicine using mathematical modelling tools from the bioinformatics field. Mathematical modelling of cellular systems uses simplifications to infer meaningful representations of the unknown gene regulatory network (GRN), and generates new testable hypotheses. My research focuses on simplifications that can perform genome-wide modelling by combining several layers of epigenetic information to derive T-cell specific genome-wide GRNs. In this talk I will present two projects which exploit ordinary differential equations (ODEs) to describe the expression dynamics of most genes. The resulting networks are functionally validated and also explored within the context of preventative and personalized medicine. The first project exploits the popular LASSO approach combining time-series expression dynamics, predicted transcription factor (TF) bindings and methylation to infer a GRN. From the network we identified upstream key TFs, and in combination with genome-wide association studies we identified asymptomatic regulators of two relapsing T cell associated diseases, namely seasonal allergic rhinitis and multiple sclerosis. The second project uses nonlinear ODE modelling of a core system of TFs within differentiation and its dynamical impact on thousands of genes.