A Tool for Synapse-level Circuit Analysis of Human Cerebral Cortex Specimens.

项目概要/摘要 这项工作的目标是促进神经元及其互连微电路的突触水平分析， 神经外科大脑皮层活检标本。这些全层人类大脑皮层活检 将由接受切除手术或手术植入的患者的神经外科同事提供 脑深部电刺激（DBS）的导线。一旦我们证明了这些技术和工具 足够可靠，我们将分析来自研究精神病的医学中心的样本中的神经回路， 神经系统疾病在初始阶段，我们将：1）优化未受损脑活检的切除， 神经外科手术和转移新技术的浸泡固定和锇染色，以大（>5 从病人身上提取的新鲜的大脑活组织切片。2)我们将优化硬件和软件， 改变，速度和可靠性的多束扫描电子显微镜图像采集自动 在脑容量中获得PB级的突触级神经回路， 神经元通过数以亿计的突触; 3）我们将转移方法来共同注册分子标签（用于细胞 - 类型），使用非常小的新型免疫探针对相同的人类样品进行连续电子显微镜检查， 不需要透化，因此不会对大脑超微结构的质量产生负面影响， 以确定每种细胞类型的超微结构相关性; 4）我们将与计算机科学家合作， 阿贡国家实验室开发一个强大的计算连接管道，用于拼接，对齐， 分割和存储人脑回路。这个公共平台将增强团队的努力， 谷歌已经开始研究我们的人体样本。在第二阶段，我们将运行许多人类 通过图像采集和分析连接管道进行活检。我们将使用新的软件来比较 个体内部和个体之间的回路变异性。我们认为对人脑神经回路的详细分析 将纳米尺度连接到毫米尺度的组织是理解正常组织如何 大脑功能和发现认知和发育障碍的病理基础。我们 我们的目标是传播我们开发的方法，成为神经病理学工具箱的一部分， 和基本的人类神经科学。

Project Summary/Abstract The goal of this work is to facilitate synaptic level analysis of neurons and their interconnecting microcircuits in neurosurgical cerebral cortex biopsies from human patients. These full-thickness human cerebral cortex biopsies will be provided by neurosurgical colleagues from patients undergoing resective surgery or surgical implantation of leads for deep brain stimulation (DBS). Once we have demonstrated that the techniques and tools are sufficiently reliable, we will analyze neural circuits in samples from medical centers that study psychiatric and neurological disorders. In the initial phase we will: 1) optimize the removal of undamaged brain biopsies during neurosurgical procedures and transfer new techniques for immersion fixation and osmium staining to large (>5 cubic millimeter) fresh brain biopsies from human patients.; 2) we will optimize, with hardware and software changes, the speed and reliability of multibeam scanning electron microscopy image acquisition to automatically acquire synapse-level neural circuitry at petabyte scale in brain volumes that connect tens of thousands of neurons via hundreds of millions of synapses; 3) we will transfer methods to co-register molecular labels (for cell -types) with serial electron microscopy of the same human samples using very small novel immuno-probes that do not require permeabilization and hence, do not negatively impact the quality of the brain’s ultrastructure in order to identify ultrastructural correlates for each cell type; and 4) we will work with computer scientists at Argonne National Laboratory to develop a robust computational connectomics pipeline for stitching, alignment, segmentation, and storage of human brain circuits. This public platform will augment the efforts of a team at Google that is already begun working on our human samples. In the second phase, we will run many human biopsies through the image acquisition and analysis connectomic pipelines. We will use new software to compare circuit variability within and between individuals. We contend that detailed neural circuit analysis in human brain tissue that bridges scales from nanometers to millimeters is a prerequisite for understanding how the normal brain functions and discovering the pathological underpinnings of cognitive and developmental disorders. Our goal is that the methods we develop will be disseminated, becoming part of the toolbox for both neuropathology and fundamental human neuroscience.