Predicting the spread and impact of transmissible vaccines

通常在野生动物中流行的传染病偶尔也会进入人类。例如，狂犬病病毒每年感染并杀死数万人，因为携带病毒的野生动物咬伤人类并将病毒传播给他们。其他偶尔从动物传染给人类的病毒甚至更危险，因为它们也可以在人与人之间传播，从而可能引发流行病或大流行病。不幸的是，我们尚未制定有效的解决方案来阻止这些传染病蔓延到人群中。相反，我们目前对这些动物疾病的方法是反应性的，并侧重于对已感染的人类进行医学治疗，并在人类疫情蔓延并成为全面流行病或大流行病之前将其控制起来。解决这一挑战性问题的一个有希望的解决方案是开发可以在动物之间传播的野生动物疫苗。通过自我传播，这些疫苗扩大了野生动物群体内的免疫力传播，并减少或消除了向人类群体蔓延的风险。虽然正在开发多种自我传播的动物疫苗，但我们还没有数学，统计和计算工具，我们需要批判性地评估它们的性能，从而对其可能的使用做出明智的决定。该项目的工作将开发这些定量工具，并使候选的自我传播疫苗在使用前得到严格的评价。此外，该项目将培训来自农村背景的第一代大学生使用数学和计算模型来评估和优化对美国经济未来至关重要的新兴生物技术。学生招聘将通过提供有竞争力的财政支持来促进，以减轻放弃研究经验而选择传统就业的压力。最后，本项目将继续开发网站，向公众介绍自传播疫苗，传播相关研究成果，并审查这一新兴技术的现状。在决定进行小规模的现场试验之前，应量化自传播疫苗改善人类健康的可能性。这一要求带来了巨大的技术挑战，因为在放行前无法收集疫苗在靶动物群体中的行为数据。该项目将利用重组载体传播疫苗的数学模型克服这一技术挑战，该模型可结合实地和实验室数据进行参数化。具体而言，将开发数学模型，整合水库人口的年龄结构和明确的模式，疫苗脱落从动物感染疫苗。这些模型将采用偏微分方程系统的形式。现场数据将来自对储存动物的诱捕研究，记录每只捕获动物的年龄以及是否被用于构建候选疫苗的载体病毒感染。实验室数据将描述实验感染疫苗的储毒动物的疫苗脱落时间模式。近似贝叶斯计算将被用来参数化的模型和随机模拟框架预测的结果，拟议的疫苗发布。通过反复模拟疫苗释放的模型参数化随机绘制后验分布，这个框架忠实地集成了水库生态，生物过程中的随机性，参数估计的不确定性。该项目开发的方法将应用于拉沙病毒的原型自传播疫苗，但将广泛适用于为一系列动物宿主开发的自传播疫苗。该项目由环境生物学部人口和社区生态学（PCE）集群，刺激竞争研究的既定计划（EPSCoR），该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Infectious diseases that normally thrive in wild animals occasionally make the leap into the human population. For instance, rabies virus infects and kills tens of thousands of people each year when wild animals carrying the virus bite humans and transmit the virus to them. Other viruses that occasionally leap from animals to humans are even more dangerous because they can also transmit from human to human and thus potentially seed epidemics or pandemics. Unfortunately, we do not yet have effective solutions in place to stop these infectious diseases from spilling over into the human population. Instead, our current approach to these animal diseases is reactive, and focuses on medical treatment of humans who have become infected and corralling human outbreaks before they can spread and become full blown epidemics or pandemics. A promising solution to this challenging problem is the development of wildlife vaccines that can spread themselves from one animal to the next. By self-disseminating, these vaccines magnify the spread of immunity within the wild animal population and reduce or eliminate the risk of spillover into the human population. Although multiple self-disseminating animal vaccines are being developed, we do not yet have the mathematical, statistical, and computational tools we need to critically evaluate their performance and thus make informed decisions about their possible use. Work on this project will develop these quantitative tools and enable candidate self-disseminating vaccines to be critically evaluated before they are used. In addition, this project will train first-generation college students from rural backgrounds to use mathematical and computational models to evaluate and optimize emerging biotechnologies critical to the future of the US economy. Student recruitment will be facilitated by offering competitive financial support that relieves pressure to abandon research experiences in favor of traditional employment. Finally, this project will continue development of a website that explains self-disseminating vaccines to the public, disseminates relevant research results, and examines the state of this emerging technology.Before making the decision to conduct even small-scale field trials, the likelihood that a self-disseminating vaccine will improve human health should be quantified. This requirement poses a formidable technical challenge because data on the behavior of the vaccine within the target animal population cannot be collected prior to release. This project will overcome this technical challenge using mathematical models of recombinant vector transmissible vaccines that can be parameterized using a combination of field and laboratory data. Specifically, mathematical models will be developed that integrate the age structure of the reservoir population and the explicit pattern of vaccine shedding from animals infected with vaccine. These models will take the form of a system of partial differential equations. Field data will come from trapping studies of the reservoir animal that record the age of each captured animal and whether it was infected by the vector virus used to construct the candidate vaccine. Laboratory data will describe the temporal pattern of vaccine shedding from reservoir animals experimentally infected with the vaccine. Approximate Bayesian computation will be used to parameterize the models and a stochastic simulation framework developed for predicting the outcome of a proposed vaccine release. By repeatedly simulating a vaccine release for models parameterized by drawing randomly from the posterior distribution, this framework faithfully integrates reservoir ecology, randomness in biological processes, and uncertainty in parameter estimates. The methodology developed by this project will be applied to a prototype self-disseminating vaccine for Lassa virus but will be broadly applicable to self-disseminating vaccines developing for a range of animal reservoirs.This project is jointly funded by the Population and Community Ecology (PCE) Cluster in the Division of Environmental Biology, the Established Program to Stimulate Competitive Research (EPSCoR), and the Mathematical Biology Program in the Division of Mathematical and Physical Sciences.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.