Parallel Software for Fast, Automated Determination of Virus Structures

项目负责人/主要研究者（最后一名、第一名、中间名）：Baker，Timothy 1 R 01 AI 079095 -01 项目摘要 现代电子冷冻显微镜和基于计算机的3D图像重建技术正在彻底改变大型复杂生物大分子机器结构的研究方式。这些方法提供了在分子水平上理解这些机器如何运作的关键。科学 我们研究的目标是在尽可能高的分辨率和最短的时间内可靠地确定病毒结构，因为这将提高我们了解病毒如何引起各种疾病的能力。 这种结构工作可能会产生关于如何最好地开发用于预防病毒感染的抗病毒药物的重要线索。显微镜和图像重建各自提出了一系列重大挑战，我们强大、高效的软件的持续开发显然是我们和其他人研究工作成功的关键。该提案旨在大幅提高计算能力，包括我们的冷冻重建工作的3D结构确定部分的心脏。 这项工作的最终目标是在显微镜下近实时地直接进行二十面体病毒的3D图像重建。在接下来的两年里，我们实际上希望在收集显微照片后不到五分钟内达到低分辨率（25- 30 A）。这将是完成 通过提高数字密集型应用程序的速度和并行可扩展性，增强我们的自动图像重建系统，并修改现有的图像装箱软件，以便在没有人工干预的情况下识别单个病毒颗粒。我们还计划 当没有合适的起始图时，提高了用于从未对准图像重新构造起始模型的随机模型方法的可靠性。即使在重建不是在显微镜下完成的情况下，这项工作也将大大简化和加速结构确定项目。在 结合上述努力，我们将实施算法更改，使其有可能达到更高的分辨率。 这些新的能力预计将对许多生物学问题产生深远的影响，因为它们允许人们快速筛选样品的所需特性。我们实验室目前的挑战包括确定P22突变体中Nanogold标记的半胱氨酸残基的位置， 狒狒呼肠孤病毒中5重对称轴附近的“瓣”，抗体和其它配体与人细小病毒表面的结合化学计量，以及由于甲病毒的E2糖蛋白中的小（19个残基）插入引起的总体形态学变化。拟议的工作在很大程度上受到了同事们在冷冻显微镜，图像处理，数据可视化和解释，最近收购了两个新的显微镜与美国国立卫生研究院的支持，并与圣地亚哥超级计算机中心的互动的密切联系。我们所有的软件都将随时提供给电子显微镜社区。 PH 39812590（修订版11/07）第2页续格式第页

Program Director/Principal Investigator (Last, First, Middle): Baker, Timothy 1R01 AI079095-01 Project Summary Modern electron cryo-microscopy and computer-based, 3D image reconstruction techniques are revolutionizing the way structures of large and complex biomacromolecular machines are studied. These methods provide keys to understanding at the molecular level how these machines function. The scientific goal of our studies is to determine virus structures reliably at the highest possible resolutions and in the shortest amount of time, as this will enhance our ability to understand how viruses cause a variety of diseases. Such structural work may yield important clues about how best to develop anti-virals for preventing viral infections. Microscopy and image reconstruction each pose their own set of significant challenges, and continued development of our robust, efficient software clearly represents a key to the success of our research efforts as well as that of others. This proposal aims to substantially enhance the computational capabilities that comprise the heart of the 3D structure determination part of our cryo-reconstruction work. The ultimate goal of the proposed work is to perform 3D image reconstructions of icosahedral viruses directly at the microscope in near real-time. Over the next two years, we realistically expect to reach low resolutions (25-30A) in less than five minutes after a micrograph has been collected. This will be accomplished through improvements to the speed and parallel scalability of the numerically intensive applications, enhancements to our automated image reconstruction system, and modifications to existing image boxing software so that individual virus particles can be identified without manual intervention. We also plan to improve the reliability of the random model method for constructing a starting model de novo from unaligned images when a suitable starting map is not available. Even in instances when the reconstructions are not done at the microscope, this work will greatly streamline and accelerate structure determination projects. In conjunction with the efforts described above, we will implement algorithmic changes that make it possible to reach higher resolutions. These new capabilities are expected to have a profound impact on many problems of biological interest since they allow one to rapidly screen samples for desired properties. Current challenges in our lab include determining the locations of Nanogold-labeled cysteine residues in P22 mutants, the presence or absence of "flaps" in the vicinity of 5-fold symmetry axes in baboon reovirus, binding stoichiometry of antibodies and other ligands to the surface of human parvoviruses, and gross morphological changes due to smalw I 9 residue) insertions in the E2 glycoprotein of alphaviruses. The proposed work is heavily leveraged by close ties to colleagues in cryo-microscopy, image processing, and data visualization and interpretation, the recent acquisition of two new microscopes with NIH support, and interactions with the San Diego Supercomputer Center. All of our software will be made readily accessible to the electron microscopy community. PHs 39812590 (Rev. 11/07) Page 2 Continuation Format Page