Gregoire Altan-Bonnet ImmunoDynamics Group Programs in Computational Biology & Immunology Memorial Sloan-Kettering Cancer Center, New York NY Title: Modeling how immune responses get reliably established despite unreliable lymphocytes Abstract: Decision making in the immune system generally implies large-scale coordination of lymphocytes’ activity over varied spatio-temporal scales. Our previous work demonstrated how unreliable the activation of isolated T lymphocytes can be. In that context, cell-cell communications are critical to proofread the response of individual T cells. One prominent mechanism to enforce such cellular communication is via cytokines, i.e. small proteins that lymphocytes can secrete and respond to. These cytokines have been shown to be critical to maintain homeostasis, to enforce peripheral tolerance, to coordinate differentiation at the global level, to match the diversity of potential pathogenic challenges (viral, bacterial, fungal, etc). Here we present a quantitative model of interleukin-2 (IL-2) communication between T cells. We find that the characteristic accumulation of IL-2 scales with the strength of antigen activation, but is made independent of T cell precursor frequency via complex feedback regulation. We also demonstrate quantitatively how competition for IL-2 between effector T cells and regulatory T cells can be a major mechanism to decide between immune response and tolerance. We will discuss how computer models (with experimental validation) are critical to demonstrate how communication via cytokine regulates T cell activation. Specifically, we will discuss how multi-scale agent-based approaches (from individual cell to populations) are necessary to probe the dynamics of the immune system quantitatively.