QSAR study Molecular docking of matrix metalloproteinases inhibitory activity of hydroxamate derivatives by MIA-QSAR using OSC-GA-PLS

The quantitative structure-activity relationship (QSAR) is an important part of computer aided drug design. the design and model programs of a QSAR analysis base on MIA-QSAR analysis were presented. In this QSAR study, the compounds of matrix metalloproteinase inhibitors (MMP-2 and MMP-9) on anticancer activity were investigated by various chemometrics methods. Predicting its anti-cancer activity in this method is particular importance. The detailed application of the multivariate image analysis (MIA) method to the evaluation of a quantitative relationship between molecular structure and inhibitory activity of hydroxamate derivatives as inhibitors of matrix metalloproteinase (MMP-2 and MMP-9) as anticancer agents was discovered. MIA is a type of data mining method based on data sets obtained from 2D images (descriptors). The purpose of this research is to construct a relationship between pixels of images of studied compounds as independent variables and their inhibitory activities as a dependent variable. The resulted descriptors were exposed to principal component analysis (PCA) [1,2]. The pixel descriptors have been used for modeling using MIA and the effect of correcting vertical and coupled signals by genetic algorithm has been discussed. the modeling stage have been compared using the partial least squares (PLS) methods and the orthogonal signal correction (OSC) method was combined with the partial least squares (PLS) method. the results of PLS, GA- PLS, OSC- PLS, OSC-GA- PLS methods have been compared with using statistical results. the resultant OSC-GA-PLS model had a high statistical quality (R2=0.98) for predicting the inhibitory activity of the compounds [3]. MIA-QSAR (multivariate image analysis-quantitative structure activity relationship) proved to be a highly predictive approach. It also showed that the OSC-GA-PLS method is better than the traditional PLS method. It can also be used to predict the inhibitory activity of new compounds. Finally, molecular docking was performed for the selected compounds in QSAR with the appropriate receptor and acceptable results were obtained. These results are suitable for predicting compounds with better properties.