The influence of the topology of regulatory networks on the distribution of avalanches in gene expression data.

This paper describes the effects of perturbations, whichsimulate the knock-out of single genes, one at a time, in randomBoolean models of genetic networks (RBNs). The analysis concentrates on the probability distribution of so-called avalanches (defined in the text) in gene expression. The topology of the random Boolean networks considered here is of the scale-free type, with a power-law distribution of outgoing connectivities. The results for these scale-free randomBoolean networks (SFRBNs) are firstly compared with those ofclassical RBNs, which had been previously analyzed, secondly withexperimental data on S. cerevisiae. It is shown that, while both models approximate the main features of the distribution of experimental data, SFRBNs tend to overestimate the number of large avalanches