STRUCTURAL BIOLOGY--IMAGE PROCESSING OF ELECTRON MICROGRAPHS

本项目利用图像处理技术分析电子 显微照片 为了回答结构生物学中的重要问题， 必须获得相对高分辨率二和三， 有关生物大分子的维度结构信息。 而 原子或近原子分辨率信息传统上 可通过X射线晶体学用于一些小分子和小 蛋白质，绝大多数生物大分子不是 晶体，或者太大，因此不适合3D 结晶学 另一方面，生物样本可以在电子显微镜中可视化。 显微镜使用一些标本制备技术。 负 染色和阴影，都使用重金属，是两个传统的 增加对比度以显示生物大分子的方法 结构 冷冻电子显微镜，一种新的技术，试图 通过在标本周围覆盖一层 冰 与LSBR、NIAMS的合作研究目前正在进行， 许多这样的项目，其中电子显微图像是 计算校正、组合、平均、重构，或在一些实施例中， 以计算方式增强，以提高信噪比或 增加被可视化的结构的可解释性。 cryo 图像的对比度通常较低，需要更高的计算机处理能力 以达到满意的效果。 我们的研究特别感兴趣的是对病毒的理解， 结构. 目前，我们正在继续努力调查 一种大型动物病毒，人类单纯疱疹病毒（1型）。 我们 正在确定主要衣壳的位置 proteins. 利用三维二十面体重建 技术，我们应用这些病毒粒子的对称性， 随机取向的衣壳颗粒的取向（在冰中）和联合收割机 许多粒子到一个三维重建。 生物 这些疱疹重建的材料是通过合作提供的， 弗吉尼亚大学夏洛茨维尔分校的研究人员合作。 的 在LSBR、NIAMS中进行电子显微镜检查。 解读我们的3D 重建由所有合作者共同执行。

This project uses image processing techniques to analyze electron micrographs. In order to answer important questions in structural biology, it is necessary to obtain relatively high resolution two- and three- dimensional structural information about biological macromolecules. While atomic or near atomic resolution information has traditionally been available by x-ray crystallography for some small molecules and small proteins, the overwhelming majority of biological macromolecules are not crystalline, or are too large and therefore not amenable to 3-D crystallography. Biological specimens can, on the other hand, be visualized in the electron microscope using a number of specimen preparation techniques. Negative staining and shadowing, which both use heavy metals, are two traditional approaches to increasing contrast to show the biological macromolecule's structure. Cryo-electron microscopy, a newer technique, attempts to preserve "native" structure by surrounding the specimen with a layer of ice. Collaborative studies with LSBR, NIAMS are currently underway on a number of such projects, whereby the electron micrograph images are computationally corrected, combined, averaged, reconstructed, or in some way computationally enhanced to improve the signal-to-noise ratio or to increase the interpretability of the structures being visualized. Cryo images are typically lower contrast and require greater computer processing to achieve satisfactory results. Of particular interest to our research is the understanding of viral structures. At present we are continuing our efforts to investigate the structure of a large animal virus, human herpes simplex virus (type 1). We are in the process of determining the location of the major capsid proteins. Using the three-dimensional icosahedral reconstruction technique, we apply the symmetry of these virus particles both to find the orientation of randomly oriented capsid particles (in ice) and to combine many particles into a three-dimensional reconstruction. Biological material for these herpes reconstructions is provided through collaboration with researchers at the University of Virginia, Charlottesville. The electron microscopy is performed in LSBR, NIAMS. Interpretation of our 3-D reconstructions is performed jointly by all collaborators.