COMPUTER RESOURCE--MOLECULAR MECHANICS AND NEURO-ANATOMY

计算机资源——分子力学和神经解剖学

• 批准号: 3103645
• 负责人: CYRUS LEVINTHAL
• 金额: $ 77.63万
• 依托单位: COLUMBIA UNIV NEW YORK MORNINGSIDE
• 依托单位国家: 美国
• 财政年份: 1976
• 资助国家: 美国
• 起止时间: 1976-08-01 至 1988-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/3103645
• 关键词: brain mapping; cell cell interaction; computer center; computer graphics /printing; computer simulation; computer system design /evaluation; conformation; image processing; molecular biology; monitoring device; neuroanatomy; neurons

There are three major aspects of the proposed Biotechnology Research Resource. First, the completion, programming and utilization of a new computing machine (Molecular Mechanics Computing Machine, MMCM) specifically designed to carry out the calculations involved in theoretical approaches to protein structure and function. The new system, which is currently being constructed, will provide computing power which is between 500 and 1,000 times greater than a VAX 11/780, the computer most commonly used for this work. It will be significantly faster than any of the super-computers now available or anticipated in the near future. However, it is very specific for problems in which most of the computing time is spent in determining the forces on each of many atoms which interact with pairwise central forces. These force calculations are to be carried out on a new hard-wired processor which we call FASTRUN and the calculation of the atomic displacement will be done on a commercial array processor. After the first year or so of the proposed grant when the system is functioning it will be available for collaborative and service use. The system will be used to study the interactions of enzymes and substrates, receptors and hormones and drugs designed as inhibitors of overactive proteins. The second major activity of the facility is the further development and use of an Automatic Nerve Tracing system for the 3-D reconstruction of portions of the neuropil of the brains of small mammals and simple invertebrates. New applications of the progress in computer science and image analysis will be applied to the problem and the system will be specifically designed to handle the EM images produced in various collaborative projects. The scientific problems are primarily those related to developmental neurobiology and the factors which influence the normal growth and connectivity pattern of brains. The third aspect of the work of the facility would involve the further development and use of our CARTOS (Computer Aided Reconstruction from Serial Sections) LOANER systems. These systems provide equipment and programs for tracing neurons in 3-D which can be used by investigators who are not located in the New York area. The LOANERS would be upgraded to provide for partial automation, both for EM work and work directly on light microscopes. One very important new collaborative project which will use the LOANERs is one in which a 3-D rat brain atlas will be maintained in the computer and programs will be written to allow neurobiologists working on neural-peptide localization to enter data directly into a 3-D data base and relate the new data to the atlas and other peptide localizations.

拟议的生物技术研究有三个主要方面 资源。 首先，新的完成，编程和利用 计算机（分子力学计算机，MMCM） 专门设计用于执行理论中涉及的计算 蛋白质结构和功能的方法。 新系统，即 目前正在构建，将提供计算能力 比VAX 11/780高500和1,000倍，这是计算机最常见的 用于这项工作。 它的速度明显比任何一个 超级计算机现在在不久的将来可用或预期。 然而， 对于大多数计算时间的问题，它非常具体 花在确定与与之相互作用的每个原子上的力 成对中央力。 这些力量计算将在 我们称为Fastrun的新的硬连线处理器和计算 原子位移将在商业阵列处理器上进行。 后 系统运行时提议的赠款的第一年左右 它将用于协作和服务使用。 系统将是 用于研究酶和底物，受体和底物的相互作用 激素和药物设计为过度活跃蛋白的抑制剂。 该设施的第二个主要活动是进一步的发展和 使用自动神经追踪系统进行3-D重建 小型哺乳动物和简单的大脑神经胶质的一部分 无脊椎动物。 计算机科学进展的新应用以及 图像分析将应用于问题，系统将是 专门设计用于处理各种产生的EM图像 协作项目。 科学问题主要是 与发育神经生物学以及影响的因素有关 大脑的正常生长和连通性模式。 该设施工作的第三个方面将进一步 开发和使用我们的Cartos（计算机辅助重建 串行部分）借贷系统。 这些系统提供设备， 追踪3D神经元的程序，调查人员可以使用 不在纽约地区。 贷款人将被升级为 提供部分自动化，无论是用于EM工作还是直接工作 显微镜。 将使用的一个非常重要的新协作项目 借贷者是将3-D老鼠大脑地图集保持在 计算机和程序将被编写以允许神经生物学家从事 神经肽定位将数据直接输入3-D数据库，然后 将新数据与地图集和其他肽定位相关联。