QSAR MODELING OF GELATINASE INHIBITORS: MLR APPROACH

Abstract

Matrix metalloproteinases are regulated by growth factors, hormones, cytokines etc. and endogenous inhibitors control them. Over expression of matrix metalloproteinases can lead to a variety of pathological disorders which results because of imbalance between the activity of matrix metalloproteinase inhibitors and tissue inhibitors of matrix metalloproteinase. So, it is the necessity to develop the Phenylalanine analogues with impact inhibitive concentration. Here Phenylalanine analogues have been used to correlate the inhibiting activity with the Eccentric Connectivity index (ECI), Fragment Complexity (FC) and Topological Polar Surface Area (TPSA) for studying the Quantitative Structure Activity Relationship (QSAR). Correlation may be an adequate predictive model which can help to provide guidance in designing and subsequently yielding greatly specific compounds that may have reduced side effects and improved pharmacological activities. We have used Multiple Linear Regression (MLR), one of the best methods for developing the QSAR model. Results from this QSAR study have suggested that ECI, FC and TPSA are the important descriptors for inhibitory activities of endopeptidase inhibitors. For the validation of the developed QSAR model, statistical analysis such as data point-descriptor ratio, fraction of variance, cross validation test, standard deviation, quality factor, fischers test; and internal validation such as Y-randomization test have been performed and all the tests validated this QSAR model.