Abstract: The dynamics of infectious diseases are complex phenomena that unfold across many spatial scales. Contact patterns between individuals determine a disease's prevalence and duration in small populations, while the structure of mobility and transportation networks determines the way epidemics may spread across large distances. With the availability of high performance computing, high-resolution computer simulations have become one of the most popular strategies for making predictions in outbreak scenarios like the recent Ebola crisis in West Africa and related emergent infectious diseases. These simulations are often preferred because they account for as much detail as available data permits. However, in acute outbreak situations, data on initial conditions and plausible values for model parameters are typically scarce, and predictions made by even the most sophisticated computational models are often not robust. Furthermore, the internal complexity of models can mask the universal dynamical features that may be hidden in observed spreading processes. Using methods from network science, one can ask which features of the observed spreading patterns are determined by structural properties of the network. Based on this idea, I will show that global disease dynamics that are typically characterized by spatially incoherent patterns can actually be equated with simple, wave-like patterns observed in ordinary reaction diffusion systems if a different concept of distance is used.
If time permits, I will also discuss properties of disease dynamics on temporal networks and high-resolution contact networks and the challenges involved in understanding disease dynamics on a smaller scale.