Computational Chemistry, Lead Identification and Optimization Support, Boehringer Ingelheim Pharma GmbH & Co. KG, 88397 Biberach, Germany.
J Comput Chem. 2011 Jun;32(8):1743-52. doi: 10.1002/jcc.21758. Epub 2011 Mar 4.
Factor Xa (fXa) is a promising target for antithrombotic drugs. Recently, we presented a molecular dynamics study on fXa, which highlighted the need for a careful system setup to obtain stable simulations. Here, we show that these simulations can be used to predict the free energy of binding of several fXa inhibitors. We tested molecular mechanics/Poisson-Boltzmann surface area, molecular mechanics/Generalized Born surface area, and linear interaction energy (LIE) on a small data set of fXa ligands. The continuum solvent approaches only yield satisfying correlations to the experimental results if some of the water molecules are explicitly included in the free energy calculations. LIE gave reasonable results if a sufficiently large data set is used. In general, our procedure of setting up the fXa simulation system enabled MD simulations, which produce adequate ensembles for free energy calculations.
Xa 因子（fXa）是一种有前途的抗血栓药物靶点。最近，我们对 fXa 进行了一项分子动力学研究，该研究强调需要仔细设置系统以获得稳定的模拟。在这里，我们表明这些模拟可用于预测几种 fXa 抑制剂的结合自由能。我们在 fXa 配体的小数据集上测试了分子力学/泊松-玻尔兹曼表面区域、分子力学/广义 Born 表面区域和线性相互作用能（LIE）。连续溶剂方法只有在某些水分子被明确包含在自由能计算中时，才能与实验结果产生令人满意的相关性。如果使用足够大的数据集，LIE 会给出合理的结果。总的来说，我们设置 fXa 模拟系统的程序使 MD 模拟能够产生足够的自由能计算的集合。
