Richardson Jane S, Schneider Bohdan, Murray Laura W, Kapral Gary J, Immormino Robert M, Headd Jeffrey J, Richardson David C, Ham Daniela, Hershkovits Eli, Williams Loren Dean, Keating Kevin S, Pyle Anna Marie, Micallef David, Westbrook John, Berman Helen M
Department of Biochemistry, Duke University Medical Center, Durham, North Carolina, 27710-3711, USA.
RNA. 2008 Mar;14(3):465-81. doi: 10.1261/rna.657708. Epub 2008 Jan 11.
A consensus classification and nomenclature are defined for RNA backbone structure using all of the backbone torsion angles. By a consensus of several independent analysis methods, 46 discrete conformers are identified as suitably clustered in a quality-filtered, multidimensional dihedral angle distribution. Most of these conformers represent identifiable features or roles within RNA structures. The conformers are given two-character names that reflect the seven-angle delta epsilon zeta alpha beta gamma delta combinations empirically found favorable for the sugar-to-sugar "suite" unit within which the angle correlations are strongest (e.g., 1a for A-form, 5z for the start of S-motifs). Since the half-nucleotides are specified by a number for delta epsilon zeta and a lowercase letter for alpha beta gamma delta, this modular system can also be parsed to describe traditional nucleotide units (e.g., a1) or the dinucleotides (e.g., a1a1) that are especially useful at the level of crystallographic map fitting. This nomenclature can also be written as a string with two-character suite names between the uppercase letters of the base sequence (N1aG1gN1aR1aA1cN1a for a GNRA tetraloop), facilitating bioinformatic comparisons. Cluster means, standard deviations, coordinates, and examples are made available, as well as the Suitename software that assigns suite conformer names and conformer match quality (suiteness) from atomic coordinates. The RNA Ontology Consortium will combine this new backbone system with others that define base pairs, base-stacking, and hydrogen-bond relationships to provide a full description of RNA structural motifs.
利用所有的主链扭转角,为RNA主链结构定义了一种共识分类和命名法。通过几种独立分析方法的共识,在经过质量过滤的多维二面角分布中,46个离散构象被确定为适当聚类。这些构象中的大多数代表了RNA结构中可识别的特征或作用。这些构象被赋予两个字符的名称,反映了经验上发现有利于糖-糖“套件”单元的七个角度(δ、ε、ζ、α、β、γ、δ)组合,其中角度相关性最强(例如,A-型为1a,S-基序起始为5z)。由于半核苷酸由δ、ε、ζ的数字和α、β、γ、δ的小写字母指定,这个模块化系统也可以解析来描述传统的核苷酸单元(例如,a1)或二核苷酸(例如,a1a1),这在晶体学图谱拟合水平上特别有用。这种命名法也可以写成一个字符串,在碱基序列的大写字母之间有两个字符的套件名称(例如,GNRA四环为N1aG1gN1aR1aA1cN1a),便于生物信息学比较。提供了聚类均值、标准差、坐标和示例,以及从原子坐标分配套件构象名称和构象匹配质量(适合度)的Suitename软件。RNA本体联盟将把这个新的主链系统与其他定义碱基对、碱基堆积和氢键关系的系统结合起来,以全面描述RNA结构基序。