A platform for stochastic modeling of malaria epidemiology and control

The complexities of the Plasmodium life-cycle and of the dynamics of individual infections impose limitations on the applicability of conventional deterministic models of malaria epidemiology. We have now constructed an individual-based stochastic simulation model to predict the potential impact of interventions on malaria epidemiology. This model incorporates effects of host age on exposure to mosquito bites, of the success probability of inoculations as a function of EIR; and of naturally acquired immunity on parasite densities and consequent malaria morbidity and mortality. We have optimised estimates of the model parameters using multiple field datasets and a statistical fitting algorithm. Enormous computational power is needed to fit such models, to carry out sensitivity analysis, and to explore the potential impact of a wide range of different intervention strategies. We obtain this using a volunteer computing system, whereby volunteers offer spare computer capacity over the internet. This system can provide a general platform for comparing, fitting, and evaluation of a range of different models of malaria epidemiology.