Carosati Emanuele, Sciabola Simone, Cruciani Gabriele
Laboratory for Chemometrics and Cheminformatics, Department of Chemistry, University of Perugia, Via Elce di Sotto 10, I-06123 Perugia, Italy.
J Med Chem. 2004 Oct 7;47(21):5114-25. doi: 10.1021/jm0498349.
Through the years the GRID force field has been tuned to fit experimental observations in crystal structures. This paper describes the determination of the hydrogen bonding pattern for organic fluorines based on an exhaustive inspection of the Protein Data Bank. All the PDB complexes, whose protein structures have cocrystallized fluorine-containing ligands, were examined and geometrically inspected. By applying statistics, the hydrogen bonding geometry was described as a distribution function of the angle at the fluorine: a new specific angular function was consequently defined and inserted in the program GRID to estimate the effect of fluorine hydrogen bonds on the ligand-protein binding. All the fluorine-containing ligands collected from the PDB were docked within their corresponding protein binding sites: introducing the fluorine hydrogen bonding contribution improves the results of the docking experiments in terms of accuracy and ranking.
多年来,GRID力场已经过调整,以符合晶体结构中的实验观察结果。本文描述了基于对蛋白质数据库的详尽检查来确定有机氟的氢键模式。检查并从几何角度考察了所有其蛋白质结构与含氟配体共结晶的PDB复合物。通过应用统计学方法,将氢键几何结构描述为氟原子处角度的分布函数:因此定义了一个新的特定角度函数,并将其插入到GRID程序中,以估计氟氢键对配体 - 蛋白质结合的影响。从PDB收集的所有含氟配体都对接至其相应的蛋白质结合位点内:引入氟氢键贡献在准确性和排序方面改善了对接实验的结果。