Wu C.H., McLarty J.W. Neural Networks and Genome Informatics

Elsevier, 2000, -219 p.The resurgence of interest in artificial neural networks fortunately coincided with the emergence of new technology in molecular biology and the explosion of information about the genomes of humans and other species. Many important problems in genome informatics have been successfully addressed with artificial neural networks, and a vast literature has developed within the last two decades. The purpose of this book is to introduce molecular biologists and other informatics scientists to artificial neural network technology and terminology; to review the major neural network applications in genome informatics; to address the important issues in applying neural network technology to informatics; and to identify significant remaining problems.
Part I of this book gives an overview of applications of artificial neural network technology. Part II contains a tutorial introduction to the most commonly used neural network architectures, network training methods, and applications and limitations of the different architectures. Part III reviews the current state of the art of neural network applications to genome informatics and discusses crucial issues such as input variable selection and preprocessing. Finally, Part IV identifies some of the remaining issues and future directions for research, including integration of statistical rigor into neural network applications, hybrid systems and knowledge extraction.Part I Overview
Neural Networks for Genome Informatics
Part II Neural Network Foundations
Neural Network Basics
Perceptrons and Multilayer Perceptrons
Other Common Architectures
Training of Neural Networks
Part III Genome Informatics Applications
Design Issues - Feature Presentation
Design Issues - Data Encoding
Design Issues - Neural Networks
Applications - Nucleic Acid Sequence Analysis
Applications - Protein Structure Prediction
Applications - Protein Sequence Analysis
Part IV Open Problems and Future Directions
Integration of Statistical Methods into Neural Network Applications
Future of Genome Informatics Applications