Molecular modelling methods for prediction of sequence-selectivity in DNA recognition
Molecular modelling methods for prediction of sequence-selectivity in DNA recognition
复制标题
DOI：
10.1016/j.ymeth.2006.09.002
复制
发表时间：
2007-06-01
期刊：
影响因子：
4.8
通讯作者：
Laughton, Charles A.
中科院分区：
文献类型：
--
作者：
Wang, Hao;Laughton, Charles A.
我们描述了如何应用基于分子力学/广义 Born (MM/GB) 方法的分子建模方法来预测 DNA 小沟结合配体对不同 DNA 序列的相对亲和力。我们讨论了该技术的理论背景、可采用的方法的一些变化，并通过案例研究说明了其应用：分析 Hoechst 33258 与各种富含 A/T 的 DNA 双链体小沟的结合能量。我们展示了如何建立基础分子动力学（MD）模拟，如何分析它们以获得令人满意的行为，以及从中提取药物结合热力学的各种方法。我们发现，虽然对基本 MM/GB 方法的某些改进可以提高与实验数据的一致性（例如，计算 DNA 扰动能量），但其他方法必须更加谨慎地分析（例如，计算构型熵变化）。总体而言，这些方法可以很好地对配体对不同DNA序列的小沟的亲和力进行排名，但绝对结合亲和力的计算不是很可靠。 (c) 2006 Elsevier Inc. 保留所有权利。
We describe how one can apply molecular modelling methods, based on the molecular mechanics/generalised Born (MM/GB) approach, to the prediction of the relative affinity of DNA minor groove binding ligands for different DNA sequences. We discuss the theoretical background to the technique, some variations in the methodology that can be employed, and illustrate its application through a case study: analysis of the energetics of binding of Hoechst 33258 to the minor groove of various A/T-rich DNA duplexes. We show how the underpinning molecular dynamics (MD) simulations can be set up, how they can be analysed for satisfactory behaviour, and various approaches to extracting thermodynamics of drug binding from them. We find that while certain elaborations to the basic MM/GB method can improve the agreement with experimental data (e.g., calculating the DNA perturbation energy), others have to be analysed with more caution (e.g., calculating configurational entropy changes). Overall, these methodologies can rank the affinity of a ligand for the minor groove of different DNA sequences fairly well, but the calculation of absolute binding affinities is not very reliable. (c) 2006 Elsevier Inc. All rights reserved.