Inferring systemic inflammatory dynamics from time-series gene expression

In healthy organisms, systemic inflammation enables the body to clear an infection when it is exposed to a pathogen. In sepsis, however, the global inflammatory response leads to life-threatening organ dysfunction. Many questions remain regarding how the amplification and damping of these responses are controlled in the complex, whole organism context. To trace the dynamics of these responses, we analyze time-series gene expression data from the exposure of mice to pathogen-associated molecules. Using dimensional reduction approaches, we extract collective variables that control inflammatory dynamics and generate hypotheses for corresponding biological modules. We then employ inference methods to describe the interactions between these collective variables. This approach predicts the timecourse of inflammatory states and is a step towards quantitative control of inflammation.