3D-QSAR Modeling of Substituted Thiophene-Anthranilamides as Potent Inhibitors of Human Factor XA Using Quantum Chemical Descriptors

K. Dguigui, M. Elhallaoui

Abstract: The potent inhibitors of human factor Xa of 54 thiophene derivatives were modeled by quantitative structure-activity relationship (QSAR) using density functional theory (DFT) to generate quantum descriptors. From the pool of descriptors chosen to generate the QSAR model, four descriptors are selected by multiple linear regression MLR method with a correlation coefficient RRLM= 0, 95. The predictive ability of the proposed model was assessed using neural network RNN = 0, 98 and validated by internal leave one out cross validation RCV=0, 90.

Keywords: factors Xa inhibitor, anticoagulant, QSAR, MLR, Neural Network, Cross Validation