Master equation for the degree distribution of a Duplication and Divergence network

Abstract

Network growth as described by the Duplication-Divergence model proposes a simple general idea for the evolution dynamics of natural networks. In particular it is an alternative to the well known Barabási-Albert model when applied to protein-protein interaction networks. In this work we derive a master equation for the node degree distribution of networks growing via Duplication and Divergence and we obtain an expression for the total number of links and for the degree distribution as a function of the number of nodes. Using algebra tools we investigate the degree distribution asymptotic behavior. Analytic results show that the network nodes average degree converges if the total mutation rate is greater than 0.5 and diverges otherwise. Treating original and duplicated node mutation rates as independent parameters has no effect on this result. However, difference in these parameters results in a slower rate of convergence and in different degree distributions. The more different these parameters are, the denser the tail of the distribution. We compare the solutions obtained with simulated networks. These results are in good agreement with the expected values from the derived expressions. The method developed is a robust tool to investigate other models for network growing dynamics.

Evolving Network Modeling Driven by the Degree Increase and Decrease Mechanism

“…Network topological properties are also worth studying. There are a lot of traditional methods of calculating degree distributions of networks, e.g., mean-field method and master equations, which are applied under some specific situations to derive degree distributions of evolving networks [16] [17]. However, in this case of evolving networks with growth and deletion of nodes, it is difficult to obtain an analytical solution due to the non-homogeneity.…”

A comprehensive statistical study of metabolic and protein–protein interaction network properties

“…Though a single metabolic and a single protein-protein interaction (PPI) network are analyzed in the paper, these data sets composed of single elements are not enough to extract strong general conclusions about these biological systems. In [20] a version of the duplication and divergence model is analytically studied and it is observed that although it is possible to find parameter regions where the degree distribution will have a denser right tail, this distribution has a peak for low degrees indicating that it can not be fitted to a monotonically decreasing power-law function if one considers its whole range.…”

Evolution of coagulation-fragmentation stochastic processes using accurate chemical master equation approach

“…Hockin-Mann reaction network model and classic Becker-Döring-type models do not incorporate discreteness and stochasticity of the CF process, when it happens in confined space [22,13]. However, the Chemical Master Equation (CME) approach is widely used to address discreteness and stochasticity [23,24,25]. Solving the CME provides an evolving landscape in state space while the discrete form of the CME (dCME) can account for finite size effects [26,27,28].…”