RII Track-2 FEC: Marshalling Diverse Big Data Streams to Understand Risk of Tick-Borne Diseases in the Great Plains

蜱传疾病，包括莱姆病、落基山斑疹热和其他疾病，越来越被认为是世界范围内一个重大的公共卫生问题，在大平原地区尤其受到越来越多的关注。然而，对这些疾病的详细了解，它们是如何获得的，高风险地区在哪里，以及如何最好地减轻它们，仍然令人惊讶地不透明。该项目由堪萨斯大学、堪萨斯州立大学、匹兹堡州立大学、俄克拉何马州立大学、俄克拉何马大学、俄克拉何马大学诺曼校区和俄克拉何马中央大学合作，代表了一个范围广泛、高度跨学科、综合和数据密集的努力，以阐明大平原上堪萨斯州和俄克拉何马州两个州的这些问题。该项目的主要内容包括收集详细的大规模数据集，包括不同蜱虫物种的出现情况、蜱虫和病原体的基因组以及整个地区的环境变化，以及编组新的人工智能工具，以便非专业人员能够快速准确地识别蜱虫。项目科学家将使用生态位建模和数学种群建模方法来评估和预测主要蜱传病原体的传播，并创建和测试自动识别工具。该项目将通过一系列研讨会和课程以及在线数据资源来培养所谓的“大数据素养”。也许最重要的是，该项目将涉及众多本科生和研究生的许多项目任务，让他们有机会学习和探索这些和相关科学领域的未来。项目学生将尽可能广泛地招募，以代表在科学领域没有充分代表的特定人群，包括少数民族、妇女和那些来自没有大学教育传统的家庭的人。项目成果将包括蜱传疾病风险的在线互动地图，以及用于识别公众拍摄的蜱虫照片的在线设施。项目团队的初级成员（年轻的教师、博士后和学生）将得到更资深的个人的指导和指导，以最大限度地提高他们在该领域成功发展的可能性。项目小组将由一个咨询委员会指导，该委员会具有广泛的国际专业知识以及州一级的公共卫生政策经验。在项目结束时，我们期望对堪萨斯州和俄克拉何马州蜱传疾病的多样性和风险有更详细的了解。在美国，蜱传疾病越来越被认为是一个重要的公共卫生问题，包括莱姆病、埃利希体病、落基山斑疹热、南方蜱相关的皮疹病、人类粒细胞无形体病、巴贝斯虫病以及哈特兰病毒和波旁病毒感染。对蜱和病原体物种的空间分布以及相关的蜱传疾病在大平原传播的空间风险的了解相当有限。该项目汇集了几个“大数据”流（蜱虫发生数据、蜱虫和病原体基因组数据、表征环境的遥感数据）和新的科学工具，以阐明导致这些疾病的病原体感染风险的地理模式和时间动态。具体来说，这个项目是堪萨斯大学、堪萨斯州立大学、匹兹堡州立大学、俄克拉荷马州立大学、俄克拉荷马州大学、俄克拉荷马州大学诺曼校区和俄克拉荷马州中央大学之间的合作项目，将涉及从堪萨斯州和俄克拉荷马州生态不同地区的12个地点收集蜱虫和脊椎动物，并使用基因组工具对多种病原体进行测试；蜱虫和病原体分布的结果数据将成为使用互补的尖端工具（相关生态位模型，机制种群模型）详细建模传播风险的基础，以实现种群和范围动态的新综合。该项目将探索和开发基于深度学习方法的第一个自动蜱虫识别系统，该系统将以详细分布数据的形式为该项目设想的其他分析提供更多信息。该项目还将对扩大科学参与产生重大影响，通过联合、协作和综合的努力，将高级和初级科学家（包括少数教员）联系起来，以指导和建立初级团队成员（教员、博士后、研究生、本科生）以及以教学为重点的机构的教员之间的信心和地位。该项目将在大数据分析、大数据访问和疾病风险绘图等领域提供各种教育机会，这些机会将广泛和公开地提供给科学界。最后，为了减轻蜱传疾病对南部大平原社区的影响，该项目将制作并广泛提供该地区每种蜱传疾病的详细风险地图，并创建自动识别蜱虫的技术，以便公民科学家和公众利益相关者更好地获取有关蜱虫和疾病风险的个人和直接相关的信息。该奖项反映了美国国家科学基金会的法定使命，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。

Tick-borne diseases, including Lyme disease, Rocky Mountain spotted fever, and others, are increasingly appreciated as a significant public health concern worldwide, and an increasing concern in the Great Plains in particular. However, a detailed understanding of these diseases, how they are acquired, where are the high-risk areas, and how might they best be mitigated, has remained surprisingly opaque. This project, a collaboration between University of Kansas, Kansas State University, Pittsburgh State University, Oklahoma State University, the University of Oklahoma, the University of Oklahoma, Norman campus, and the University of Central Oklahoma, represents a broad-scope, highly interdisciplinary, integrated, and data-intensive effort to illuminate these questions across two states, Kansas and Oklahoma in the Great Plains. Major elements of the project include assembling detailed, large-scale datasets on the occurrences of different tick species, the genomes of the ticks and the pathogens, and environmental variation across the region, as well as marshaling new artificial-intelligence tools to permit rapid and accurate tick identifications by non-experts. Project scientists will use ecological niche modeling and mathematical population modeling approaches to assess and predict transmission of the major tick-borne pathogens, and create and test the automated identification tools. The project will foster what can be termed "big data literacy" via a series of workshops and courses, as well as online data resources. Perhaps most importantly, the project will involve numerous undergraduate and graduate students in many project tasks, giving them opportunities to learn and explore futures in these and related areas of science. Project students will be recruited as broadly as possible, to represent in particular populations that are not well-represented in science, including minorities, women, and those from families without a tradition of university-level education. Project outcomes will include online, interactive maps of tick-borne disease risk, and online facilities for identification of tick photographs taken by the general public. Junior members of the project team (younger faculty, postdocs, and students) will be mentored and guided by more senior individuals, so as to maximize the probability of their successful advancement in this field. The project team will be guided by an advisory board with broad and international expertise, as well as state-level public health policy experience. At the close of the project, we anticipate a much-improved and considerably more detailed understanding of the diversity and risk of tick-borne diseases across Kansas and Oklahoma.Tick-borne diseases are increasingly recognized as an important public health concern across the United States, including Lyme disease, ehrlichiosis, Rocky Mountain spotted fever, southern tick-associated rash illness, human granulocytic anaplasmosis, babesiosis, and viral infections with Heartland and Bourbon viruses. Knowledge of the spatial distributions of ticks and pathogen species, and the associated spatial risk of transmission of tick-borne diseases in the Great Plains is quite limited. This project marshals several "big data" streams (tick occurrence data, tick and pathogen genomic data, remote sensing data to characterize environments) and novel scientific tools to shed light on geographic patterns and temporal dynamics of risk of infection with the pathogens that cause these diseases. Specifically, this project, a collaboration between University of Kansas, Kansas State University, Pittsburgh State University, Oklahoma State University, the University of Oklahoma, the University of Oklahoma, Norman campus, and the University of Central Oklahoma, will involve field collections of ticks and vertebrates hosting ticks from 12 sites in ecologically distinct regions across Kansas and Oklahoma, which will be tested using genomic tools for a diverse suite of pathogens; the resulting data on tick and pathogen distributions will be the basis for detailed modeling of transmission risk using complementary, cutting-edge tools (correlative niche models, mechanistic population models) to achieve new syntheses of population and range dynamics. This project will explore and develop a first automated tick identification system based on deep-learning approaches, which will feed much more information into the other analyses envisioned for this project in the form of detailed distributional data. This project will also have substantial implications for broadening participation in science, via linking senior and junior scientists (including minority faculty members) in a joint, collaborative, and integrative effort designed to mentor and build confidence and stature among junior team members (faculty, postdocs, graduate students, undergraduate students) and among faculty from teaching-focused institutions. This project will offer various educational opportunities in the areas of big data analytics, big data access, and disease risk mapping, which will be made broadly and openly available to the scientific community. Finally, towards mitigating effects of tick-borne diseases in communities across the southern Great Plains, this project will produce and make broadly available detailed risk maps for each tick-borne disease in the region, and create technology for automating tick identifications, to allow citizen scientists and stakeholders in the general public better access to information on ticks and disease risks that are personally and immediately relevant.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.