IIBR Multidisciplinary: RUI: Cross Departmental Development of an Automated Species Identification System for the Phylum Tardigrada found on Birds.

迟滞，也被称为水熊，包括一种微小的水生动物门，以其通过经历隐生物作用来抵御极端环境条件和环境快速变化的能力而闻名，这种过程使它们能够与环境一起干燥，并在接近绝对零度的温度、6000大气压和高真空中生存下来。最近，水熊因其在外层空间生存的能力而受到全球媒体的关注。虽然迟滞的基本分子生物学和生理学在器官移植和对皮肤紫外线辐射的洞察中产生了令人兴奋的应用，但人们对这一群体的全球多样性和分布知之甚少。迟滞在世界各地都有发现，从海洋底部到山顶，但人们对哪些物种在哪里以及它是如何到达那里的知之甚少。该项目旨在揭示迟滞的扩散、分布和多样性。将开发一个自动机器人系统来隔离、保存、标记实验室载玻片上的迟滞样本并对其进行成像，以便永久存储。然后，这些图像被用来开发和“训练”人工智能软件来识别物种。新的、稀有的和常见的形态种的潜在发现和描述将扩大关于表型多样性的区域知识。在一个主要博物馆开发新的区域阿尔法分类凭单，全世界都可以使用，这是物种鉴定的宝贵工具，将促进将迟滞纳入全球环境和生物地理分析。准备好的标本和标本准备和鉴定的自动化系统存储在加州大学戴维斯分校的博哈特博物馆，将来将供世界各地的学生、研究人员和博物馆使用。为此，来自不同种族、性别和能力背景的高素质学生开始在贝克大学湿地贝克教师的指导下实地工作，从当地森林栖息地及其栖息地的鸟类种群中收集样本。这是一个为期三年的项目，旨在通过在Bohart博物馆(加州大学戴维斯分校)的支持下，在一所小型本科学校(贝克大学)的三个STEM系(生物学、物理学和计算机科学系)之间的合作，使研究缓和门表型生物多样性所需的两个基本基础设施过程现代化。贝克大学的教职员工监督本科生研讨会，通过设计和构建旨在减少时间、劳动力和成本的显微镜载玻片准备、标记和成像系统，首先自动化已有100多年历史的制作标本载片和图像的过程，以识别最近在鸟类上发现的迟滞动物的生物多样性。其次，教师和学生研究人员通过开发图像/模式匹配软件，利用人工智能为学生、研究人员和博物馆建立的新图像参考库，实现标本与物种表型识别的自动化。这项研究的产品将产生学生和教职员工PI撰写的论文、章节和文章。这些技术的设计和使用说明将被公布，以便大学和博物馆的学生、研究人员和团队可以建立自己的系统。该项目将证明，收藏管理、桌面机器人技术、人工智能和标本计算机识别是对无脊椎动物表型多样性和生态学进行严格科学研究所需的一些现代化基础设施领域。最后，这项对鸟类迟滞生活的多维生态学研究的结果将使生命之树的黑暗区域得到光明。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Tardigrades, also known as water bears, comprise a microscopic aquatic animal Phylum famous for their ability to withstand extreme environmental conditions and rapid changes in their environment by undergoing cryptobiosis, a process that allows them to desiccate with their environment and survive exposure to temperatures of near absolute zero, 6,000 atmospheres of pressure, and high vacuums. Recently water bears have gained global media attention for their ability to survive exposure to outer space. While the underlying molecular biology and physiology of tardigrades produced exciting applications in organ transplants and insights into ultra-violet radiation on the skin, little is known about the global diversity and distribution of this group. Tardigrades are found all over the world, from the bottom of oceans to the tops of mountains, but little is known about which species is where and how it got there. The project aims to unfold tardigrade dispersion, distribution, and diversity. An automated robotic system to sequester, preserve, label, and image specimen tardigrades on laboratory slides for permanent storage will be developed. These images are then used to develop and “train” artificial intelligence software to identify species. The potential discovery and description of new, rare, and common morphospecies will expand regional knowledge of phenotypic diversity. The development of a new regional alpha taxonomic voucher collection at a major museum will be accessible worldwide, an invaluable tool in species identification which will promote the inclusion of tardigrades in global environmental and biogeographical analyses. The prepared specimens and the automated system for preparation and identification of specimens, stored at the Bohart Museum at the University of California, Davis, will be available for students, researchers, and museums world-wide to access in the future. To this end, highly qualified students from diverse ethnic, gender, and ability backgrounds begin by working in the field under the mentorship of Baker faculty at the Baker University Wetlands to collect samples from the local forested habitats and their resident bird populations. This is a three-year project to modernize two of the fundamental infrastructure processes necessary for research into the phenotypic biodiversity of Phylum Tardigrada by a collaboration among three STEM departments (Biology, Physics & Computer Science) at a small undergraduate school (Baker University) with support from the Bohart Museum (University of California, Davis). Baker University faculty supervise undergraduate workshops to identify the biodiversity of tardigrades recently discovered on birds by first automating the 100+ yr-old process used to make slides and images of specimens by designing and building a microscope slide preparation, labeling and imaging system intended to reduce time, labor and cost. Second, faculty and student researchers automate the phenotypic identification of specimens to species by developing image/pattern matching software using a new image reference library built with artificial intelligence for students, researchers and museums. The products of this research will yield papers, chapters, and articles written by the students and faculty PIs. The design and directions for use of these technologies will be published such that students, researchers, and teams at universities and museums can build their own systems. This project will demonstrate that collections management, tabletop robotics, artificial intelligence, and computer identification of specimens are some of the infrastructure arenas requiring modernization for rigorous scientific research into invertebrate phenotypic diversity and ecology. Finally, the results of this multi-dimensional ecological study into the lives of tardigrades on birds will spread light onto a dark area of the tree of life. (http/: “mywaterbears.com”)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.