Multi-resolution Approaches to Modeling the 3D Structure, Delivery, and Replication of Viral Genomes

本项目将开发病毒感染定量研究的计算方法。在这些的核心 方法是核酸和蛋白质的多分辨率描述，允许混合分辨率 模拟非常大的生物分子系统，精确的分辨率从粗到细的切换，反之亦然， 包括一个完全原子化的表示，和一个明确的机制来解释生化转化。 基于最近的DNA多分辨率模型，该项目将开发一种计算方法， 确定加压和自组装病毒衣壳内病毒基因组的物理组织。的 方法将被应用于解决几个包装基因组的结构，在一个适合于药物的分辨率 开发应用程序。与此同时，将建立细菌和真核细胞质的多分辨率模型。 发展到占特异性和非特异性相互作用的细胞质蛋白与双链 DNA.该模型将被应用于确定双链基因组的空间组织弹射到 细胞质，并评估对喷射过程的细胞质样环境的影响。多- 分辨率模拟框架将阐明控制基因组喷射的微观因素和 完整病毒颗粒通过核孔复合物的运输。最后，该项目将开发第一个 病毒基因组复制的物理模型，解释了基本的生化转化和影响 外力对反应速率的影响。复制模型将用于确定 宿主细胞的DNA结合蛋白影响病毒基因组复制保真度。多分辨率模拟 通过该计划开发的方法将在GPU加速代码原子分辨率中实现 布朗动力学方法和代码，沿着所有必需的文档、示例和教程， 将免费提供给研究界，以研究广泛的生物物理过程。

This project will develop computational approaches for quantitative studies of viral infection. At the core of these approaches is a multi-resolution description of nucleic acids and proteins that permits mixed-resolution simulations of very large biomolecular systems, accurate resolution switching from coarse to fine and vice versa, including a fully atomistic representation, and an explicit mechanism to account for biochemical transformations. Building on a recent multi-resolution model of DNA, the project will develop a computational method for determining the physical organization of viral genomes inside pressurized and self-assembled viral capsids. The method will be applied to resolve the structure of several packaged genomes at a resolution suitable for drug development applications. In parallel, a multi-resolution model of bacterial and eukaryotic cytoplasm will be developed to account for specific and nonspecific interactions of the cytoplasmic proteins with double-stranded DNA. The model will be applied to determine the spatial organization of double-stranded genomes ejected into cytoplasm and to evaluate the effect of the cytoplasm-like environment on the ejection process. The multi- resolution simulation framework will elucidate the microscopic factors governing genome ejection and the transport of an intact viral particle through a nuclear pore complex. Finally, the project will develop the first physical model of a viral genome replication, accounting for essential biochemical transformations and the effect of external forces on the reaction rates. The replication model will be used to determine how competition between DNA binding proteins of the host cell affect viral genome replication fidelity. The multi-resolution simulation methods developed through this program will be implemented in a GPU-accelerated code Atomic Resolution Brownian Dynamics. The methods and the code, along with all required documentation, examples and tutorials, will be made freely available to the research community to study a wide range of biophysical processes.