Predicting Epidemic Risk from Past Temporal Contact Data
(A), (B): Probability distributions of the infections potentials for loyal (πL, green) and disloyal nodes (πD, orange), for the cattle trade network and the sexual contact network, respectively. Loyalty is set with a threshold ϵ = 0.1. Dashed lines show the fit with a Landau+exponential model (see Material and Methods). (C), (D): Loyalty transition probabilities between loyal statuses (TLL(k), green) and disloyal statuses (TDD(k), orange) as functions of the degree k of the node, for the cattle trade network and the sexual contact network, respectively. Dashed lines represent the logarithmic models: TDD(k) = 0.78−0.11log k, and TLL(k) = 0.63+0.06log k for the cattle trade network; TDD(k) = 0.94−0.10log k, and TLL(k) = 0.25+0.17log k for the sexual contact network. Transition probabilities are computed as frequencies in the datasets under study. The error bars here represent one binomial standard deviation from these frequencies. In (C) the error bars are smaller than the size of the points. A single pair of configurations is considered here as example; the behavior observed is the same for all the pair of configurations.