TRD2: Phasing and refinement

TRD 2.分阶段和细化-Penczek（负责人） 总结 与X射线晶体学类似，微晶电子衍射（MicroED）提供成像的振幅。 对象傅里叶变换和相位信息丢失。在X射线中已经开发了几种程序 晶体学用于解决从头结构确定的相位问题。其中包括帕特森 差异，分子置换，从头算统计方法，重金属（多晶型和单晶型） 替换）、基于损伤的定相和异常色散。我们已经成功地将MicroED应用于 用两种方法确定新的蛋白质结构：（1）直接从头算和分子置换 使用理想化模型和（2）使用同源物的分子置换。然而，从头算方法是 仅在获得的分辨率优于~1.2 μ m的情况下才可行，而分子置换仅 如果新的蛋白质与另一种已知的结构同源，则是可能的。其他X射线方法是 不适用于MicroED（例如异常色散）或尚未适应和实施 微型ED。在这里，我们将通过开发专用的MicroED定相方法来扩展可用工具的托盘 对于衍射点未达到空间频率的情况下的常规从头结构测定 从头算方法所需的，并且由于缺乏同源结构， 信息.为了促进分阶段，我们将开发全面的MicroED数据扩展和集成 利用最大似然方法来提供准确的强度值的方法。所以德 我们将采用两种在X射线晶体学中历史上最成功的方法：（1）重定相 金属同晶置换策略和（2）通过特定辐射损伤分阶段。目标是：1. 开发全面的MicroED数据缩放和集成方法; 2.从头定相，使用 同晶置换; 3.使用辐射损伤的从头定相。成功完成目标 将使MicroED与X射线晶体学平起平坐。从长远来看， 发展将导致建立微ED作为一种方法，能够分阶段和解决结构的 这是一个全新的、具有重要生物学意义的系统，目前还不能用X射线晶体学来处理。

TRD 2. Phasing and refinement - Penczek (Lead) Summary Similarly as X-ray crystallography, microcrystal electron diffraction (MicroED) delivers amplitudes of the imaged object Fourier transform and the phase information is lost. Several procedures have been developed in X-ray crystallography for solving the phase problem for de novo structure determination. These include Patterson difference, molecular replacement, ab initio statistical methods, heavy metal (multi- and single isomorphous replacement), damage-based phasing, and anomalous dispersion. We already successfully applied MicroED to the determination of new protein structures using two approaches: (1) direct ab initio and molecular replacement using idealized models and (2) molecular replacement using homologues. However, ab initio methods are feasible only in cases when the obtained resolution is better than ~1.2Å while molecular replacement is only possible if the new protein is homologous with another already known structure. Other X-ray methods are either not applicable to MicroED (for example anomalous dispersion) or were not yet adapted and implemented for MicroED. Here we will expand the pallet of available tools by developing dedicated MicroED phasing methods for routine de novo structure determination in cases in which the diffraction spots do not reach spatial frequencies required by ab initio methods and molecular replacement is not possible due to lack of homologous structural information. To facilitate phasing, we will develop comprehensive MicroED data scaling and integration methodologies taking advantage of maximum likelihood approaches to deliver accurate intensity values. For de novo phasing we will adapt two methods that were historically most successful in X-ray crystallography: (1) heavy metal isomorphous replacement strategies and (2) phasing by specific radiation damage. The aims are: 1. Development of a comprehensive MicroED data scaling and integration methodology; 2. De novo phasing using isomorphous replacement; 3. De novo phasing using radiation damage. The successful completion of the Aims listed will bring MicroED to an equal footing with X-ray crystallography. In the long-run, the proposed developments will lead to establishment of MicroED as a method capable of phasing and solving structures of entirely new and biologically important systems that currently are not tractable by X-ray crystallography.