Mathematical and computational approaches for viral infection

点击翻译按钮获取中文摘要

The project, in the fields of Mathematical Virology and Mathematical Immunology, is based on the hypothesisthat mathematical models have potential to provide an alternative to animal experiments, and may be sufficiently predictive to provide evidence to support crucial decisions regarding medical treatment strategies. The main aim of the project is to develop new models for the co-evolution of EBOV and therapeutic interfering particles (TIPs) and to link these population-level models with within-host models, and with the experimental data generated by Dstl. The objectives of the project are- to develop new mathematical models for the population-level transmission of TIPs and WT EBOV, constructingthese new models based on existing WT EBOV models. The aim of these models will be to studythe co-evolution and co-transmission of WT EBOV and TIPs,- to use predictions and data regarding within-host dynamics for informing population-level models (e.g.,transmission rate among individuals as a function of within-host viral load, recovery rate of individualsdepending on the within-host immune status, or individual clinical outcome in terms of survival/death as afunction of viral load),- to identify appropriate summary statistics (stochastic descriptors) for assessing the efficacy of TIPs fordisease propagation control (e.g., reproduction number, size of the outbreak),- to compare this disease propagation control measure with alternative existing ones, and to analyse, makinguse of the epidemiological model, the effect of combined strategies, and- to use predictions from the newly-developed stochastic population-level models to inform scientists at Dstlabout the minimum multiplicity of infection (e.g., in terms of TIP competitive advantage with the WT) forTIPs to become an efficient disease propagation control measure.Novelty of the research projectThe mathematical novelty and challenge of the project is to bring together the molecular, cellular and populationscales to understand virus and infection kinetics. The student will make use of generalised birth and deathMarkov processes, the theory of stochastic descriptors [1], Bayesian inference, and agent-based modelling, sothat together with the experimental data from Dstl, he can predict the desired multiplicity of infection of WTEBOV and TIPs, for TIPs to become an efficient disease propagation control measure.The student will make use of novel matrix analytic methods to study and analyse a number of stochasticdescriptors and to study probability and times to viral extinction. The student will also make use of novelBayesian statistical methods to bring together experimental data generated at Dstl with the mathematical modelsof within-host viral infection developed in the project in order to carry out parameter inference. He will alsodevelop novel agent-based models to characterise infection kinetics(I) Potential applications of the project Dstl has recently received DARPA funding to develop novel medical treatments for Ebola virus, based on the use of therapeutic interfering particles (TIPs). The models generated in this project will support the design of combined infection spread control measures, which could include the use of TIPs and other existing strategiesfor EBOV, benefiting research organisations, public health and medical institutions. These models will alsobe used by Dstl to provide UK Government with advice about the treatment of EBOV in scenarios relevant toDefence, Security and Public Health. Finally, government mechanisms for sharing information with internationalpartners will be exploited to enable the outputs of this project to have international impact on decision-makingand research related to Public Health, Defence and Security.