Prev. Article Next Article Table of Contents

Research Article

Use of Automatic Relevance Determination in QSAR Studies Using Bayesian Neural Networks

Frank R. Burden*

School of Chemistry, Monash University, Victoria 3800, Australia

Martyn G. Ford and David C. Whitley

Centre for Molecular Design, Institute of Biomedical and Biomolecular Sciences, Portsmouth University, Portsmouth PO1 2DY, U.K.

David A. Winkler

CSIRO Division of Molecular Science, Private Bag 10, Clayton South MDC, Clayton, Victoria 3169, Australia

J. Chem. Inf. Comput. Sci., 2000, 40 (6), pp 1423–1430

DOI: 10.1021/ci000450a

Publication Date (Web): October 28, 2000

To whom correspondence should be addressed. Telephone: +613 9905 4559. Fax: +613 9905 4597. E-mail: frank.burden@sci.monash.edu.au.

Abstract

We describe the use of Bayesian regularized artificial neural networks (BRANNs) coupled with automatic relevance determination (ARD) in the development of quantitative structure−activity relationship (QSAR) models. These BRANN-ARD networks have the potential to solve a number of problems which arise in QSAR modeling such as the following: choice of model; robustness of model; choice of validation set; size of validation effort; and optimization of network architecture. The ARD method ensures that irrelevant or highly correlated indices used in the modeling are neglected as well as showing which are the most important variables in modeling the activity data. The application of the methods to QSAR of compounds active at the benzodiazepine and muscarinic receptors as well as some toxicological data of the effect of substituted benzenes on Tetetrahymena pyriformis is illustrated.

View: Full Text HTML | Hi-Res PDF

Tools

SciFinder Links

Get Reference Detail

History

Published In Issue November 27, 2000
Received March 21, 2000

Recommend & Share

Related Content

Spline-Fitting with a Genetic Algorithm: A Method for Developing Classification Structure−Activity RelationshipsJournal of Chemical Information and Computer Sciences
- Spline-Fitting with a Genetic Algorithm: A Method for Developing Classification Structure−Activity Relationships
  Jeffrey J. Sutherland, Lee A. O'Brien, and Donald F. Weaver
  Journal of Chemical Information and Computer Sciences 2003 43 (6), pp 1906–1915
  Abstract: Classification methods allow for the development of structure−activity relationship models when the target property is categorical rather than continuous. We describe a classification method which fits descriptor splines to activities, with descriptors ...
  Abstract | Full Text HTML | Hi-Res PDF
Effect of Selection of Molecular Descriptors on the Prediction of Blood−Brain Barrier Penetrating and Nonpenetrating Agents by Statistical Learning MethodsJournal of Chemical Information and Modeling
- Effect of Selection of Molecular Descriptors on the Prediction of Blood−Brain Barrier Penetrating and Nonpenetrating Agents by Statistical Learning Methods
  Hu Li, Chun Wei Yap, Choong Yong Ung, Ying Xue, Zhi Wei Cao, and Yu Zong Chen
  Journal of Chemical Information and Modeling 2005 45 (5), pp 1376–1384
  Abstract: The ability or inability of a drug to penetrate into the brain is a key consideration in drug design. Drugs for treating central nervous system (CNS) disorders need to be able to penetrate the blood−brain barrier (BBB). BBB nonpenetration is desirable for ...
  Abstract | Full Text HTML | Hi-Res PDF
Toward an Optimal Procedure for Variable Selection and QSAR Model BuildingJournal of Chemical Information and Computer Sciences
- Toward an Optimal Procedure for Variable Selection and QSAR Model Building
  A. Yasri and D. Hartsough
  Journal of Chemical Information and Computer Sciences 2001 41 (5), pp 1218–1227
  Abstract: In this work, we report the development of a novel QSAR technique combining genetic algorithms and neural networks for selecting a subset of relevant descriptors and building the optimal neural network architecture for QSAR studies. This technique uses a ...
  Abstract | Full Text HTML | Hi-Res PDF

Research Article

Use of Automatic Relevance Determination in QSAR Studies Using Bayesian Neural Networks

Abstract

Tools

SciFinder Links

History

Recommend & Share

Related Content

Spline-Fitting with a Genetic Algorithm: A Method for Developing Classification Structure−Activity Relationships

Effect of Selection of Molecular Descriptors on the Prediction of Blood−Brain Barrier Penetrating and Nonpenetrating Agents by Statistical Learning Methods

Toward an Optimal Procedure for Variable Selection and QSAR Model Building

Other ACS content by these authors:

JOURNALS

BOOKS

MAGAZINE

COMMUNITIES

DIRECTORIES & DATABASES