Computational Infrastructure for Brain Research: EAGER: A Unified Computational Framework for Analysis, Storage, and Visualization of 3D Brain Microscopy Data

我们对神经系统功能的理解在很大程度上依赖于可视化大脑的三维结构。大脑由许多类型的细胞组成，这些细胞被组织成复杂的网络，以产生认知等神经功能。与健康人的大脑相比，神经精神疾病患者的大脑经常显示出无序的细胞放置。因此，确定细胞类型和组织的细节如何与认知和行为相关是至关重要的。结构脑成像的最新进展使以细胞分辨率创建完整大脑标本的数字数据集成为可能。然而，考虑到哺乳动物大脑中大量的神经元，用这些技术产生的数据集如此之大，以至于阻碍了研究进程。在这个项目中，我们将提出一个新的计算基础设施框架来应对这一挑战，最终目的是促进实验室之间的合作，这些实验室生成并使用这些大型细胞成像数据集来进行神经科学发现。因此，该项目符合美国国家科学基金会的使命，即促进科学进步，促进国民健康、繁荣和福利。组织透明技术和光片显微镜的最新创新使人们能够在荧光标记的大脑样本中快速获取三维微米分辨率图像。然而，由此产生的信息密集型数据集的巨大规模给跨多个实验室以标准化方式共享、分析和可视化这些数据带来了巨大的计算挑战。在该项目中，将利用与北卡罗来纳大学教堂山分校相关的三个相互关联和互补的中心的综合专业知识，通过开发可供神经科学界利用的统一且高度可扩展的计算基础设施框架来解决这一问题。几个目标是开发用于分布式存储、共享和分析高维图像的网络基础设施；开发用于3D显微图像的定量分析的高通量计算工具；提供使用沉浸式环境高效可视化结果的手段；并通过将这些工具应用于分析、共享和可视化自闭症小鼠模型的脑结构缺陷来展示这些工具的实用性。这项由CEISE高级网络基础设施司颁发的早期概念探索性研究(AGER)奖由CEISE信息和智能系统司联合支持，资金与NSF了解大脑、大脑倡议活动和开发国家神经科学研究基础设施有关。该项目还与国家战略计算倡议下的国家科学基金会的目标保持一致。

Our understanding of nervous system function is critically dependent on visualizing the three-dimensional structure of the brain. The brain is composed of many cell types that are organized into complex networks to produce neural functions such as cognition. The brains of patients with neuropsychiatric illness often display disorganized cell placement as compared to brains of healthy individuals. It is therefore fundamentally important to determine how the details of cell type and organization relate to cognition and behavior. Recent advances in structural brain imaging have enabled the possibility for creating digital data sets of complete intact brain specimens at cellular resolution. However, given the enormous number of neurons in the mammalian brain, the data sets produced with these techniques are so large as to hamper the research process. In this project, we will present a novel computational infrastructure framework that addresses this challenge, with the ultimate aim of facilitating collaboration among laboratories that generate and use these large cellular imaging data sets for neuroscience discovery. This project therefore aligns with the NSF mission to promote the progress of science and to advance the national health, prosperity and welfare. Recent innovations in tissue clearing techniques and light sheet microscopy allow the rapid acquisition of three-dimensional micron resolution images in fluorescently labeled brain samples. However, the enormous size of the resulting information-dense data sets present great computational challenges to sharing, analysis, and visualization of these data in a standardized manner across multiple laboratories. In this project, the combined expertise of three connected and complementary centers associated with the University of North Carolina at Chapel Hill will be leveraged to address this issue through development of a unified and highly scalable computational infrastructure framework that can be harnessed by the neuroscience community. The several aims are to develop cyberinfrastructure for the distributed storage, sharing and analysis of high-dimensional images; develop high throughput computational tools for quantitative analysis of 3D microscopy images; provide the means for efficient visualization of results using immersive environments; and demonstrate the utility of these tools by applying them to the analysis, sharing, and visualization of brain structure deficits in an autism mouse model.This Early-concept Grants for Exploratory Research (EAGER) award by the CISE Division of Advanced Cyberinfrastructure is jointly supported by the CISE Division of Information and Intelligent Systems, with funds associated with the NSF Understanding the Brain, BRAIN Initiative activities, and developing national research infrastructure for neuroscience. This project also aligns with NSF objectives under the National Strategic Computing Initiative.