Network model with structured nodes

We present a network model in which words over a specific alphabet, called structures, are associated to each node and undirected edges are added depending on some distance measure between different structures. This model shifts the underlying principle of network generation from a purely mathematical one to an information-based one. It is shown how this model differs from the Barábasi-Albert and duplication models and how it can generate networks with topological features similar to biological networks: power law degree distribution, low average path length, clustering coefficient independent from the network size, etc. Two biological networks: S. cerevisiae gene network and E. coli protein-protein interaction network, are replicated using this model.