Self and nonseld revisited: Lessons from modelling the immune network

Abstract

In this paper we present a new model for the mechanism underlying what is traditionally known in immunology as the "selfnonself’ distinction. It tunas out that in operational terms, the distinction effected by this model of the immune system is between a sufficiently numerous set of antigens present from the start of the ontogeny of the system on the one hand, and isolated antigens first introduced after the system has reached maturity on the other. The coincidence between this "founder versus late" distinction and the traditional "somatic self-foreign pathogen" one is essentially contingent, an example of the purely opportunistic tinkering characteristic of biological organization in general. We conclude that the so-called "self-nonself’ distinction in immunology is a misleading misnomer. This raises the question as to what would genuinely count as a "self-nonself’ distinction, a fundamental question for biology in general and Artificial Life in particular.