Lifetimes of small catalytic networks (Q1337343)

The dynamics of self-replicative molecules (replicators) has been extensively analysed, following the seminal work of \textit{M. Eigen} [Naturwissenschaften 58, 465-523 (1971)], as a bridge between the theory of chemical kinetics and population genetics in the framework of Darwinian evolution. From the beginning, two well-differentiated cases were distinguished: those systems in which interactions among replicators could be neglected and those in which the internal relationships among replicators should be taken into account. In the latter, usually referred as catalytic networks, \(m\) different kinds of molecules (or species) are linked by their cross-catalytic effects on the replication of other molecules. There are extensive studies of the deterministic dynamics of these systems in the limit of infinite population. However, the deterministic analysis fails when the number of molecules of a species is so small that the description in terms of concentrations is not allowed and the internal fluctuations have to be taken into account. The most dramatic difference between the deterministic and the stochastic analysis appears in the long time stability of the system. A new molecule may only appear as the result of the replication of a molecule of the same species (either catalysed or non-catalysed by other species), and it is always possible that a fluctuation results in the extinction of a species. As the probability of extinction is positive and (within the model) there is no possible recovery from it, the only truly stationary state corresponds to the total extinction of the system. However, metastable solutions exist as a signature of the deterministic fixed points. These metastable states may have such a long lifetime that, in practice, they may be considered permanent. In this paper we address the problem in terms of the average lifetime, \(\tau\), of a catalytic network, defined as the mean time to lose in the first species of the network. \(\tau\) depends on the population size, \(N\), and the structure of the network. We focus our study on simple networks formed by small number of species, \(m \leq 5\). The main assumptions of the model and the summary of the relevant properties in the limit of infinite population are presented in sections 2 and 3, respectively. In section 4, we introduce the formulation of the problem in the case of finite populations. Section 5 is dedicated to the analysis of different few component symmetric systems. We study the role of constraints in the total population and the differences between hypercycles, fully connected networks and systems with non-catalysed replication. We find a scaling relation which links the population size and the structure of the catalytic relationships. In section 6, we present some simulation results to explore the behaviour of higher dimension replicator systems. The study of few component asymmetric systems is carried out in section 7 and the main conclusions are drawn in section 8.