QSAR

Computational models (QSAR) for estimation of volume of distribution



Computational models (QSAR) for estimation of volume of distribution

Abstract

Quantitative structure-activity relationship (QSAR) approaches were used to estimate the volume of distribution (Vd) using an artificial neural network (ANN). The data set consisted of the volume of distribution of 129 pharmacologically important compounds, i.e., benzodiazepines, barbiturates, nonsteroidal anti-inflammatory drugs (NSAIDs), tricyclic anti-depressants and some antibiotics, such as betalactams, tetracyclines and quinolones. The descriptors, which were selected by stepwise variable selection methods, were: the Moriguchi octanol-water partition coefficient; the 3D-MoRSE-signal 30, weighted by atomic van der Waals volumes; the fragment-based polar surface area; the d COMMA2 value, weighted by atomic masses; the Geary autocorrelation, weighted by the atomic Sanderson electronegativities; the 3D-MoRSE - signal 02, weighted by atomic masses, and the Geary autocorrelation - lag 5, weighted by the atomic van der Waals volumes. These descriptors were used as inputs for developing multiple linear regressions (MLR) and artificial neural network models as linear and non-linear feature mapping techniques, respectively. The standard errors in the estimation of Vd by the MLR model were: 0.104, 0.103 and 0.076 and for the ANN model: 0.029, 0.087 and 0.082 for the training, internal and external validation test, respectively. The robustness of these models were also evaluated by the leave-5-out cross validation procedure, that gives the statistics Q2 = 0.72 for the MLR model and Q2 = 0.82 for the ANN model. Moreover, the results of the Y-randomization test revealed that there were no chance correlations among the data matrix. In conclusion, the results of this study indicate the applicability of the estimation of the Vd value of drugs from their structural molecular descriptors. Furthermore, the statistics of the developed models indicate the superiority of the ANN over the MLR model.

Table of Contents

CHAPTER 1: INTRODUCTION5

Aims and Objectives7

CHAPTER 2: LITERATURE REVIEW8

A Short History Of QSAR Evolution8

Databases17

CHAPTER 3: METHODS20

Data set20

Descriptor generation and screening20

Molecular diversity analysis21

Nonlinear modeling23

Approaches to Developing a QSAR24

CHAPTER 4: RESULT AND DISCUSSION36

CHAPTER 5: CONCLUSION45

CHAPTER 1: INTRODUCTION

Reproductive and developmental toxicity studies (referred to collectively here as reprotoxicity) are used to identify the adverse effects a chemical may have on sexual function and fertility in adult males and females, developmental toxicity in the offspring, as well as effects on, or mediated via, lactation. Thus, reproductive toxicity refers to a range of endpoints relating to the impairment of male and female reproductive capacity (fertility) and the induction of non-heritable harmful effects on the progeny (developmental toxicity). The variety of observable effects are brought about by a plethora of mechanisms of action, many of which are unknown or only partially understood at the molecular and cellular level. Along with carcinogenicity studies, reprotoxicity studies are among the most costly and time-consuming experimental procedures (Lombardo, 2004, pp. 1242). Furthermore, reprotoxicity testing requires the highest number of test animals. For all these reasons, the development of alternative (non-animal) methods for reprotoxicity assessment is a high political priority.

The volume of distribution (Vd) of a drug is a major pharmacokinetic property that determines the drug half life and ...