Perturbation in genetic regulatory networks: simulation and experiments

Random boolean networks (RBN) have been proposed more thanthirty years ago as models of genetic regulatory networks. Recent studies on theperturbation in gene expression levels induced by the knock-out (i.e. silencing)of single genes have shown that simple RBN models give rise to a distributionof the size of the perturbations which is very similar in different model networkrealizations, and is also very similar to the one actually found in experimentaldata concerning a unicellular organism (S.cerevisiae). In this paper we presentfurther results, based upon the same set of experiments, concerning thecorrelation between different perturbations. We compare actual data from S.cerevisiae with the results of simulations concerning RBN models with morethan 6000 nodes, and comment on the usefulness and limitations of RBNmodels.