Suppr超能文献

一种通过熵和似然性联合最大化进行相位确定的多解方法。VI. 使用纠错码作为相位置换源及其在粉末、电子和大分子晶体学相位问题中的应用。

A multisolution method of phase determination by combined maximization of entropy and likelihood. VI. The use of error-correcting codes as a source of phase permutation and their application to the phase problem in powder, electron and macromolecular crystallography.

作者信息

Gilmore C, Dong W, Bricogne G

机构信息

Department of Chemistry, University of Glasgow, Glasgow G12 8QQ, Scotland.

出版信息

Acta Crystallogr A. 1999 Jan 1;55(Pt 1):70-83. doi: 10.1107/s0108767398007909.

Abstract

The use of error-correcting codes as a source of efficient designs of phase permutation schemes is described. Three codes are used, all taken from the Bricogne BUSTER program [Bricogne (1993). Acta Cryst. D49, 37-60]: the Hamming [7, 4, 3], the Nordström-Robinson (16, 256, 6) and the Golay [24, 12, 8] or its punctured [23, 12, 7] form. These are used in a maximum-entropy-likelihood phasing environment to carry out phase permutation of basis-set reflections instead of the usual quadrant permutation or magic integer approaches. The use of codes in this way inevitably introduces some errors in the phase choices, but for most structures this is not significant especially when the gain in sampling efficiency is considered. For example, the Golay [24, 14, 8] allows the permutation of 24 centric phases in such a way that only 4096 phase sets are produced instead of 2(24) = 16777216, and one of these sets has, at most, only four wrong phases. The method is successfully applied to three powder diffraction data sets of increasing complexity, and with increasing degrees of overlap {Mg(3)BN(3), Sigma-2 ([Si(64)O(128)].4C(10)H(17)N) and the NU-3 zeolite}, a sparse electron diffraction data set for buckminsterfullerene, C(60), and the small protein molecule crambin at 3 Å resolution where 42 reflections are phased with a Uweighted mean phase error of 58.5 degrees.

摘要

本文描述了使用纠错码作为相位排列方案高效设计的来源。使用了三种编码,均取自布里科涅的BUSTER程序[布里科涅(1993年)。《晶体学报》D49,37 - 60]:汉明[7, 4, 3]码、诺德斯特龙 - 罗宾逊(16, 256, 6)码以及戈莱[24, 12, 8]码或其删位[23, 12, 7]形式。这些编码用于最大熵似然相位确定环境中,以对基集反射进行相位排列,而非通常的象限排列或神奇整数方法。以这种方式使用编码不可避免地会在相位选择中引入一些误差,但对于大多数结构而言,这并不显著,尤其是考虑到采样效率的提高时。例如,戈莱[24, 14, 8]码允许对24个中心相位进行排列,使得仅产生4096个相位集,而不是2(24) = 16777216个,并且这些集合中最多只有四个错误相位。该方法成功应用于三个复杂度不断增加且重叠程度不断增加的粉末衍射数据集{Mg(3)BN(3)、Sigma - 2 ([Si(64)O(128)].4C(10)H(17)N)和NU - 3沸石}、一个用于巴基球C(60)的稀疏电子衍射数据集以及分辨率为3 Å的小蛋白质分子胰凝乳蛋白酶原,其中42个反射的相位通过无加权平均相位误差为58.5度进行确定。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验