Gradwell M J, Feeney J
Laboratory of Molecular Structure, National Institute for Medical Research, London, UK.
J Biomol NMR. 1996 Jan;7(1):48-58. doi: 10.1007/BF00190456.
The use of intermolecular NOEs for docking a small ligand molecule into its target protein has been investigated with the aim of determining the effectiveness and methodology of this type of NOE docking calculation. A high-resolution X-ray structure of a protein-ligand complex has been used to simulate loose distance constraints of varying degrees of quality, typical of those estimated from experimental NOE intensities. These simulated data were used to examine the effect of the number, distribution and representation of the experimental constraints on the precision and accuracy of the calculated structures. A standard simulated annealing protocol was used, as well as a more novel method based on rigid-body dynamics. The results showed some analogies with those from similar studies on complete protein NMR structure determinations, but it was found that more constraints per torsion angle are required to define docked structures of similar quality. The effectiveness of different NOE-constraint averaging methods was explored and the benefits of using 'R-6 averaging' rather than 'centre averaging' with small sets of NOE constraints were shown. The starting protein structure used in docking calculations was obtained from previous X-ray or NMR structure studies on a related complex. The effects on the calculated conformations of introducing structural differences into the binding site of the initial protein structure were also considered.
为了确定这种分子间核Overhauser效应(NOE)对接计算的有效性和方法,研究了利用分子间NOE将小配体分子对接至其靶蛋白的情况。已使用蛋白质-配体复合物的高分辨率X射线结构来模拟不同质量程度的宽松距离约束,这些约束类似于根据实验NOE强度估算得到的约束。这些模拟数据用于检验实验约束的数量、分布和表示形式对计算结构的精度和准确性的影响。使用了标准的模拟退火方案以及一种基于刚体动力学的更新颖的方法。结果显示与关于完整蛋白质NMR结构测定的类似研究有一些相似之处,但发现定义相似质量的对接结构每个扭转角需要更多的约束。探讨了不同NOE约束平均方法的有效性,并展示了在少量NOE约束情况下使用“R-6平均”而非“中心平均”的益处。对接计算中使用的起始蛋白质结构来自先前对相关复合物的X射线或NMR结构研究。还考虑了在初始蛋白质结构的结合位点引入结构差异对计算构象的影响。