Predictive Comparative QSAR Analysis Of As 5-Nitofuran-2-YL Derivatives Myco bacterium tuberculosis H37RV Inhibitors Bacterium Tuberculosis H37RV Inhibitors

摘要

Antitubercular activity of 5-nitrofuran-2-yl Derivatives series weresubjected to Quantitative Structure Activity Relationship (QSAR) Analysis withan effort to derive and understand a correlation between the biologicalactivity as response variable and different molecular descriptors asindependent variables. QSAR models are built using 40 molecular descriptordataset. Different statistical regression expressions were got using PartialLeast Squares (PLS),Multiple Linear Regression (MLR) and Principal ComponentRegression (PCR) techniques. The among these technique, Partial Least SquareRegression (PLS) technique has shown very promising result as compared to MLRtechnique A QSAR model was build by a training set of 30 molecules withcorrelation coefficient ($r^2$) of 0.8484, significant cross validatedcorrelation coefficient ($q^2$) is 0.0939, F test is 48.5187, ($r^2$) forexternal test set (pred$_r^2$) is -0.5604, coefficient of correlation ofpredicted data set (pred$_r^2se$) is 0.7252 and degree of freedom is 26 byPartial Least Squares Regression technique.