Project on Disease Network Predictability

At what level of connectivity and redundancy in dynamic heterogeneous graphs does disease carrying capacity become predictable?

Is there a point at which individual factor interventions (e.g., differences in pathogen etiology, host behavior, etc.) alter that threshold? 

Diseases are flowing across population all the time, but most of the time they are innocuous – either because they don’t travel far or because they don’t make you (too) sick. These continuous assaults only become a pandemic tipping point threat when three features coincide: (a) the connectivity of the underlying network is robust enough to allow widespread distribution at whatever (b) infectivity level the bug has, which is (c) itself dangerous.

We want to develop a model for the risk-space associated with these three thresholds. The idea is to build an epidemic simulation over complex realistic networks where we vary (a) the connectivity of the network, (b) the bug evolves on two dimensions: transmissibility and lethality, with the evolution being governed by selective pressures in the network.

The core question is to establish which sorts of control features push back the tipping point from some core baseline.