Size and connectivity as emergent properties of a developing immune network

摘要: The development of the immune repertoire during neonatal life involves a strong selection process among different clones. system is genetically capable producing much more diverse set lymphocyte receptors than are expressed in actual repertoire. By means model we investigate hypothesis that carried out early by network. We develop network which possibly hundreds B cell clones proliferate and produce antibodies following stimulation. Stimulation viewed as occurring through receptor crosslinking modeled via log bell-shaped dose-response function. Through secretion free antibody can stimulate one another if their have complementary shapes. Receptor shapes binary strings complementarity evaluated string matching algorithm. dynamic behavior our typically oscillatory for some parameters chaotic. In case two clones, chaotic attractor has number features common with Lorenz attractor. networks do not predetermined size or topology. Rather, bone marrow source generates novel These either be incorporated into remain isolated, mimicking non-network portion system. Clones removed from they fail to expand. properties function P (match), probability randomly selected immunoglobulins As evolve self-regulatory features. Most importantly, attain specific equilibrium generate characteristic amount “natural” antibody. Because reaches an asymptotic even though keeps supplying must compete presence network, i.e. takes place. Networks comprised cells multireactive small, whereas consisting become larger. find inverse relationship between n , linear M rate at produced marrow. present simple phenomenological accounts account . Additionally, suggests there qualitatively equilibria. given clone interacts with, its connectivity emergent property these networks. During ontogeny, before size, may very high. Within few months however, degree hardly dependent on receptors. appear select specificity: average always remains lower expected, favors adding over maintaining established ones. discuss “dominance” specificity, fact expected because tend occupy similar regions shape space. solution forming idiotypic complexes, parameter curve determining onset suppression, turn most crucial model. To summarize, show how could limited seemingly infinite novelty