Structural Biology and Bioinformatics Division, Indian Institute of Chemical Biology, Jadavpur, Calcutta 700032, India.
J Enzyme Inhib Med Chem. 2009 Aug;24(4):937-48. doi: 10.1080/14756360802519327.
A series of aminopyrido[2,3-d]pyrimidin-7-yl derivatives acting as potential tyrosine kinase inhibitors having anticancer activities have been considered in the present investigation for the quantitative structure-activity relationship studies based on 2D and 3D QSAR approaches. For this purpose, various theoretical molecular descriptors were computed solely from the structures of these compounds. As the number of molecular descriptors greatly exceeds the number of observations, conventional regression does not produce reliable models and therefore, ridge regression methodology was used to solve this problem. The influence of different classes of molecular descriptors on the activity has been predicted and the most significant descriptors were obtained using the ridge regression models. Partial least squares (PLS) models were developed based on the training set for the 3D QSAR models of the above compounds. The influences of steric and electrostatic field effects generated by the contribution plots are discussed.
本研究考虑了一系列作为潜在酪氨酸激酶抑制剂的氨基吡啶并[2,3-d]嘧啶-7-基衍生物，这些化合物具有抗癌活性，我们基于二维和三维定量构效关系方法进行了定量构效关系研究。为此，我们仅根据这些化合物的结构计算了各种理论分子描述符。由于分子描述符的数量大大超过观察值的数量，因此常规回归不会产生可靠的模型，因此，我们使用岭回归方法来解决这个问题。我们预测了不同类别的分子描述符对活性的影响，并使用岭回归模型获得了最显著的描述符。我们还基于训练集为上述化合物的三维定量构效关系模型开发了偏最小二乘（PLS）模型。我们讨论了由贡献图产生的立体和静电场效应的影响。
