The second edition of this highly regarded text has been substantially expanded. Part VI "Applications" is updated from 12 to 21 examples with a new focus on applications in the area of drug design. From reviews of the first edition: ?This book offers a sound introduction to artificial neuronal networks, with insights into their architecture, functioning, and applications, which is intended not only for chemists... The excellent quality of the contents and the presentation should ensure that it reaches a wide international readership.?(Angewandte Chemie) 'One of the most useful aspects of the book is a walk-through of the whole process for each application: experimental design, choice and organization of the data, selection of network architecture and parameters, and analysis of the results... The careful approach embodied in this book is an antidote to the hype which has attended neuronal networks in recent years.' (Journal of the American Chemical Society) '... highly recommended ... could become a scientific bestseller ...' (Spectroscopy Europe) 'The attractive and clear presentation of this book make it recommendable to the complete novice.' (The Analyst) 'We strongly recommend it for library purchase and it will be a useful text for lecture courses.' (Chemistry & Industry)

Jure Zupan is an author and editor of 10 books and monographs and has co-authored more than 200 articles. With Johann Gasteiger he co-authored Neural Networks in Chemistry and Drug Design. The book received more than 500 citations and was nominated the book of the month in 1993.

Part I Basic Concepts

Defining the Area

3

(6)

Learning from Information

3

(1)

General Objectives and Concepts

4

(1)

What Neural Networks are Good for

5

(2)

Notation, Conventions and Abbreviations

7

(1)

Beyond a Printed Edition

8

(1)

Neuron

9

(30)

Synapses and Input Signals

9

(2)

Weights

11

(2)

Linear Learning Machine

13

(7)

Transfer Functions in Neurons

20

(9)

Bias

29

(5)

Graphical Representation of Artificial Neurons

34

(2)

Essentials

36

(1)

References and Suggested Readings

37

(2)

Linking Neurons into Networks

39

(16)

General

39

(2)

One Layer

41

(3)

Input

44

(2)

Architectures

46

(1)

Hidden Layer; Output Layer

46

(1)

Graphical Representation of Neural Networks

47

(4)

Essentials

51

(1)

References and Suggested Readings

52

(3)

Part II One-Layer Networks

Hopfield Network

55

(12)

General

55

(1)

Architecture

56

(1)

Transfer Function

56

(1)

Weight Matrix

57

(3)

Iteration

60

(2)

Capacity of the Hopfield Network

62

(2)

Essentials

64

(1)

References and Suggested Readings

65

(2)

Adaptive Bidirectional Associative Memory (ABAM)

67

(14)

Unsupervised and Supervised Learning

67

(2)

General

69

(1)

ABAM Network

70

(1)

Learning Procedure

71

(2)

An Example

73

(3)

Significance of the Example

76

(3)

Essentials

79

(1)

References and Suggested Readings

80

(1)

Kohonen Network

81

(22)

General

81

(1)

Architecture

82

(4)

Competitive Learning

86

(5)

Mapping from Three to Two Dimensions

91

(4)

Another Example

95

(1)

Remarks

96

(2)

Essentials

98

(1)

References and Suggested Readings

99

(4)

Part III Multilayer Networks

Counter-Propagation

103

(22)

Transition from One to Two Layers

103

(1)

Lookup Table

104

(2)

Architecture

106

(1)

Supervised Competitive Learning

107

(5)

Learning to Play Tennis

112

(5)

Correlations Among the Variables

117

(3)

Essentials

120

(2)

References and Suggested Readings

122

(3)

Back-Propagation of Errors

125

(32)

General

125

(2)

Architecture

127

(1)

Learning by Back-Propagation

128

(2)

The Generalized Delta-Rule

130

(8)

Learning Algorithm

138

(3)

Example: Tennis Match

141

(11)

Essentials

152

(2)

References and Suggested Readings

154

(3)

Part IV Applications

General Comments on Chemical Applications

157

(18)

Introduction

157

(3)

Classification

160

(1)

Modeling

161

(2)

Mapping

163

(1)

Associations; Moving Window

164

(2)

Overview of the Examples in Chapters 10 to 20

166

(5)

Essentials

171

(1)

References and Suggested Readings

172

(3)

Clustering of Multi-Component Analytical Data for Olive Oils

175

(16)

The Problem

175

(1)

The Data

176

(2)

Preliminary Exploration of Possible Networks

178

(4)

Learning to Make Predictions

182

(6)

Concluding Remarks

188

(1)

References and Suggested Readings

189

(2)

The Reactivity of Chemical Bonds and the Classification of Chemical Reactions

191

(20)

The Problems and the Data

191

(3)

Architecture of the Network for Back-Propagation Learning

194

(1)

Using an Experimental Design Technique to Select the Training Set

195

(4)

Application of the Kohonen Learning

199

(1)

Application of the Trained Multilayer Network

200

(2)

Chemical Significance of the Kohonen Map

202

(2)

Classification of Reactions: The Data

204

(1)

Classification of Reactions: Results

205

(3)

References and Suggested Readings

208

(3)

HPLC Optimization of Wine Analysis

211

(8)

The Problem of Modeling

211

(1)

Modeling the Mobile Phase for HPLC by a Standard Method

212

(2)

Modeling the Mobile Phase for HPLC by a Neural Network

214

(1)

Comparison of Networks with Identical Architectures

215

(3)

References and Suggested Readings

218

(1)

Quantitative Structure-Activity Relationships

219

(24)

The Problem

219

(2)

Dataset I

221

(1)

Architecture and Learning Procedure

222

(1)

Prospects of the Method

222

(2)

Dataset II

224

(1)

Structure Representation by Autocorrelation of the Molecular Electrostatic Potential

225

(1)

Verification of Structure Representation by Unsupervised Learning

226

(2)

Modeling of Biological Activity by Supervised Learning

228

(1)

Data Set III

228

(1)

Structure Representation by Spectrum--Like Uniform Representation

229

(4)

Selection of the Most Important Variables Using a Genetic Algorithm

233

(4)

Cross-validation of the Counter-Propagation Model Obtained by the Optimal Reduced Representation

237

(4)

References and Suggested Readings

241

(2)

The Electrophilic Aromatic Substitution Reaction

243

(10)

The Problem

243

(1)

The Data

244

(1)

The Network

245

(1)

Learning and Results

246

(2)

A Third Representation of Data

248

(3)

Concluding Remarks

251

(1)

References and Suggested Readings

252

(1)

Modeling and Optimizing a Recipe for a Paint Coating

253

(8)

The Problem

253

(1)

The Data

254

(1)

The Network and Training

255

(1)

The Models

256

(3)

References and Suggested Readings

259

(2)

Fault Detection and Process Control

261

(24)

The Problems

261

(2)

The Data

263

(3)

The Methods

266

(1)

Predictions of Faults

267

(2)

Modeling and Controlling a Continuously Stirred Tank Reactor (CSTR)

269

(13)

References and Suggested Readings

282

(3)

Secondary Structure of Proteins

285

(8)

The Problem

285

(2)

Representation of Amino Acids as Input Data

287

(2)

Architecture of the Network

289

(1)

Learning and Prediction

290

(1)

References and Suggested Readings

291

(2)

Infrared Spectrum-Structure Correlation

293

(24)

The Problem

293

(2)

The Representation of Infrared Spectra as Intensities

295

(1)

The Dataset, and Learning by Back-Propagation

296

(2)

Adjustable Representation of an Infrared Spectrum

298

(1)

Representing Spectra using Truncated Sets of Fourier or Hadamard Coefficients

299

(2)

Results of Kohonen Learning

301

(5)

A Molecular Transform of the 3D Structure

306

(2)

Learning by Counter-Propagation

308

(1)

Different Strategies for the Selection of a Training Set

309

(2)

From the Infrared Spectrum to the 3D Structure

311

(2)

References and Suggested Readings

313

(4)

Properties of Molecular Surfaces

317

(20)

The Problems

317

(3)

The Network Architecture and Training

320

(3)

Tiling with Kohonen Maps; Conformational Effects

323

(2)

Investigation of Receptors of Biological Neural Networks

325

(5)

Comparison of Kohonen Maps

330

(2)

Bioisosteric Design

332

(1)

Molecular Shape Analysis

333

(2)

References and Suggested Readings

335

(2)

Libraries of Chemical Compounds

337

(10)

The Problems

337

(1)

Structure Coding

338

(2)

Separation of Benzodiazepine and Dopamine Agonists

340

(1)

Finding Active Compounds in a Large Set of Inactive Compounds

341

(1)

Diversity and Similarity of Combinatorial Libraries

342

(1)

Deconvolution of Xanthene Sublibraries

343

(3)

References and Suggested Readings

346

(1)

Representation of Chemical Structures

347

(12)

The Problem

347

(1)

Coding the Constitution

348

(2)

Coding the 3D Structure I

350

(1)

Coding the 3D Structure II

351

(4)

Coding Molecular Surfaces

355

(1)

A Hierarchy of Representations

356

(1)

References and Suggested Readings

357

(2)

Prospects of Neural Networks for Chemical Applications

359

(4)

Appendices

363

(1)

A Programs

363

(3)

B Data Sets

366

(1)

C Presentation Material

367

(1)

D Publications

367

(1)

E Tutorials

367

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.

Amazon no longer offers textbook rentals. We do!

Amazon no longer offers textbook rentals. We do!

We're the #1 textbook rental company. Let us show you why.

Neural Networks in Chemistry and Drug Design An Introduction

9783527297795

3527297790

Supplemental Materials

Summary

Author Biography

Table of Contents

Supplemental Materials

Rewards Program