Programa de Pos-Graduacao em Biologia Celular e Molecular (PPGBCM), Centro de Biotecnologia , Universidade Federal do Rio Grande do Sul (UFRGS) , Av. Bento Goncalves, 9500 , Porto Alegre , Brazil 91509-900.
Bioinformatics and High Performance Computing Research Group (BIO-HPC), Computer Engineering Department , Universidad Católica de Murcia (UCAM) , Murcia , Spain 30107.
J Chem Inf Model. 2020 Feb 24;60(2):684-699. doi: 10.1021/acs.jcim.9b00874. Epub 2020 Jan 28.
Carbohydrates are well known for their physicochemical, biological, functional, and therapeutic characteristics. Unfortunately, their chemical nature imposes severe challenges for the structural elucidation of these phenomena, impairing not only the depth of our understanding of carbohydrates but also the development of new biotechnological and therapeutic applications based on these molecules. In the recent past, the amount of structural information, obtained mainly from X-ray crystallography, has increased progressively, as well as its quality. In this context, the current work presents a global analysis of the carbohydrate information available in the Protein Data Bank (PDB). From high quality structures, it is clear that most of the data are highly concentrated on a few sets of residue types, on their monosaccharidic forms, and connected by a small diversity of glycosidic linkages. The geometries of these linkages can be mostly associated with the types of linkages instead of residues, while the level of puckering distortion was characterized, quantified, and located in a pseudorotational equilibrium landscape, not only to local minima but also to transitional states. These qualitative and quantitative analyses offer a global picture of the carbohydrate structural content in the PDB, potentially supporting the building of new models for carbohydrate-related biological phenomena at the atomistic level, including new developments on force field parameters.
碳水化合物以其物理化学、生物、功能和治疗特性而闻名。不幸的是，其化学性质给这些现象的结构阐明带来了严峻的挑战，不仅影响了我们对碳水化合物的深入理解，也影响了基于这些分子的新生物技术和治疗应用的发展。在最近的过去，主要通过 X 射线晶体学获得的结构信息量逐渐增加，其质量也有所提高。在这种情况下，目前的工作对蛋白质数据库（PDB）中可用的碳水化合物信息进行了全面分析。从高质量的结构中可以清楚地看出，大部分数据高度集中在少数几种残基类型、它们的单糖形式上，并通过少量的糖苷键连接。这些键的几何形状主要可以与键的类型而不是残基相关联，而构象的扭曲程度则被定性、定量，并定位在伪旋转平衡景观中，不仅是局部最小值，还有过渡态。这些定性和定量分析提供了 PDB 中碳水化合物结构内容的全貌，有可能支持在原子水平上构建与碳水化合物相关的生物现象的新模型，包括力场参数的新发展。
