Self Assertion Modeled as a Network Repertoire of Multi-Determinant Antibodies

We study repertoire selection in a network of natural antibodies that is maintained by simulatory idiotypic interactions. The natural antibody repertoire develops in an environment of self epitopes to which the self-reactive B cell clones are completely tolerant. e modeling formalism, we extend the shape space framework so that each antibody is represented by several randomly chosen shapes. B cell clones are generated stochastically and are removed whenever their density falls below an extinction threshold. The idiotypic interactions are governed by a log bell-shaped interaction function. tural antibody repertoire in the model is formed by percolation: the network is autonomously activated following a point stimulation. in conclusion is that the natural antibody repertoire organizes itself in such a way that most self epitopes are included in the repertoire. We find an over-representation of antibody determinants that are similar to self epitopes. We speculate the the network forms a “smoke screen” covering the somatic self.