The immune system performs a fundamental task: distinguishing between self and non-self. In an ideal scenario, formalizing this process would enable us to derive 'design principles', make accurate predictions, and create experimental interventions. In this talk, I will illustrate how starting from basic theoretical concepts, we can develop a series of models known as 'adaptive kinetic proofreading' (AKPR). These models capture key aspects of T cell recognition and shed light on immune system blindspots caused by active antagonism.

I will then discuss how we employed a specially designed robotic platform to test the core principles of adaptive kinetic proofreading in the context of collective immune responses. In a clinical context, this approach is leading us to innovative applications, such as designing a novel type of CAR-T immunotherapy bytransferring specificity from a T cell receptor to a Chimeric one . Our research demonstrates how integrating theoretical models with robotic quantitative biology can significantly enhance our understanding of cell dynamics and directly impact therapeutic development.