Abstract

The development of a computational model is reported which facilitates the study of emergent principles of human immune system effector T cell clonotype repertoire and its distribution and differentiation. In particular, the question of systemic self-organisation is addressed. The model represents an extension to earlier immune system shape space formalism, such that each activated effector T cell clonotype and respective immunogenic viral epitope is represented as a node in a two-dimensional network space, and edges between nodes models the affinity and clearance pressure applied to the antigen presenting cell bearing the target epitope. As the model is repeatedly exposed to infection by heterologous or mutating viruses, a distinct topology of the network shape space emerges which may offer a theoretical explanation of recent biological experimental results in the field of murine (mouse) cytotoxic T cell activation, apoptosis, crossreactivity, and memory - especially with respect to repeated reinfection. In the past, most discrete computational models of immune response to vira l infections have used separate real space or shape space formalisms. In this work, however, we have developed a model based on a combination of the two, with the objective of demonstrating how emergent behaviour and principles of self organisation may arise from a many-particle microscopic system. This is achieved by using a stochastic model of the lymphatic system as stimulus to a network