21-EEID Cross-scale dynamics of LASV spillover within human-driven ecosystems

在气候变化、栖息地丧失、人口增长和全球化加剧的推动下，自然界预计将在未来一个世纪发生重大变化。许多动物传播的或人畜共患的人类疾病（如埃博拉、鼠疫、炭疽）是从野生动物物种中捕获的，这些宿主物种可能会根据全球变化改变它们的行为和发现地点。最近出现的几种人畜共患疾病造成了重大的社会和经济破坏（例如SARS-COV-2、寨卡病毒、埃博拉病毒）。其中一种疾病是拉沙热，它在整个西非都有发现，每年都有广泛的爆发，每年造成数百人死亡。它是从广泛存在的农业害虫物种Mastomys natalensis中捕获的，也被称为多配偶鼠。最近的证据表明病例有所增加，因此，我们现在采取行动以更好地了解这种疾病至关重要。在这方面，我们将与人类学家密切合作，创建拉沙热病毒在动物宿主和人类人群之间传播的综合模型。我们将使用一种模型方法来检查单个啮齿动物和人类的行为，以了解宿主物种所经历的季节、地理和社会文化差异如何改变它们将病原体传播给人类的机会。我们还将包括群体行为的具体差异，例如农场工人和家庭工人。根据这些模型，科学家将提出并测试破坏人类与宿主物种之间接触的管理建议。然后，我们将使用其他更简单的方法来总结这些复杂模型的输出。这将使我们能够更多地了解我们需要了解的疾病，并在不同的空间尺度上对它们进行建模。例如，要预测一个村庄内的病例数量，我们可能需要知道大量关于个人和啮齿动物行为的信息，但要预测同样的信息，例如，在地区一级，我们可能只需要知道人和受感染的啮齿动物见面的频率。揭示疾病风险驱动因素与空间尺度之间的关系，将使我们更容易为政策制定者制作风险地图，我们知道这些地图是准确的。总的来说，使用我们不同的方法，我们可以帮助预测西非哪些地区面临拉沙热和许多其他鲜为人知的动物传播疾病的风险。通过纳入地方规模的进程，我们可以更好地制定预防疾病的措施，同时减少对贫困和脆弱人类社区生计的负面影响。此外，由于预计未来几十年西非的人口和环境将持续发生巨大变化，预测人畜共患疾病可能如何应对环境变化，以便更好地了解这些疾病未来可能在何处传播，这一点非常重要。最后，我们将创建两个软件工具。第一个项目将通过为参与式制图提供定制的数字框架，将不同学科的研究人员聚集在一起，更有效地开展工作。第二，将为未经训练的用户提供一个在线框架来运行我们的大规模疾病风险模型。

The natural world is expected to undergo a significant change over the coming century, driven by climate change, habitat loss, human population increases and increased globalisation. Many animal-borne or zoonotic human diseases (e.g. Ebola, Plague, Anthrax) are caught from wild animal species and these host species will likely alter what they do and where they are found in response to global changes. The recent emergence of several zoonotic diseases has caused significant social and economic disruption (e.g. SARS-COV-2, Zika, Ebola). One such disease is Lassa Fever, which is found throughout West Africa and has yearly annual widespread outbreaks causing hundreds of deaths a year. It is caught from the widespread, agricultural pest species Mastomys natalensis, also called the Multimammate Rat. Recent evidence has pointed to an increase in cases and, therefore, it is vital we act now to better understand this disease.In this context, working closely with anthropologists, we will create a comprehensive model of the transmission of Lassa Fever virus between the animal hosts and human populations. Using a modelling approach that examines individual rodent and human behaviour, we will look to understand how the seasonal, geographical and sociocultural differences to the conditions that host species experience, alter their chances of transmitting their pathogens to humans. We will also include specific differences in the behaviours of groups of people, such as farm labourers and household workers. From these models, will make and test management recommendations that disrupt contact between people and the host species. We will then use other, simpler methods to summarise the outputs of these complex models. This will allow us to understand more about what we need to know about diseases, to model them across different spatial scales. For instance, to predict the number of cases within a village we would likely need to know lots of information about individual human and rodent behaviour, but to predict the same information at, for instance, a district level we might just need to know how often people and infected rodents meet each other. Uncovering the relationship between drivers of disease risk and spatial scale, would allow us to more easily make risk maps for policy makers that we know are accurate.Overall, using our different approaches we can help predict which areas of West Africa are at risk of Lassa Fever and many other poorly-known animal-borne diseases. By incorporating local-scale processes we can better create measures that prevent disease, while reducing negative impacts on the livelihoods of poor and vulnerable human communities. Furthermore, with large ongoing changes to demography and the environment expected in West Africa over the coming decades, it is important to predict how zoonotic diseases will likely respond to environmental change, to better understand where these diseases may spread in the future.Lastly, we will create two software tools. The first will bring together researchers from different subjects to work more effectively, by providing a bespoke, digital framework for use in participatory mapping. The second, will provide an online framework for untrained users to run our broad-scale disease risk models.