45 CHAPTER 4 PREDICTION OF ONE-DIMENSIONAL STRUCTURAL PROPERTIES OF PROTEINS BY INTEGRATED NEURAL NETWORKS YAOQI ZHOU and ESHEL FARAGGI Indiana University School of Informatics Center for Computational Biology and Bioinformatics Indiana University School of Medicine Indiana University-Purdue University Indianapolis Indianapolis, IN Introduction to Protein Structure Prediction: Methods and Algorithms, Edited by Huzefa Rangwala and George Karypis Copyright © 2010 John Wiley & Sons, Inc. 4.1. INTRODUCTION Proteins are linear polymeric chains made of various combinations of 20 amino acid residues. The exact sequence of residues for each protein is encoded in the DNA sequence. Despite their underlying chemical simplicity, proteins can perform a wide range of biological functions from molecular signaling and transportation, molecular motors, structural support to catalyz- ing chemical reactions as enzymes. Such a multifaceted functionality is made possible by their ability to form different three-dimensional structures/shapes for different sequences of combinations of residues. A direct prediction of three-dimensional structures from protein sequences has proven challenging, as discussed in several chapters in this book. As a result, many scientists search for other structural properties that are easier to predict and are useful for aiding three-dimensional structure prediction as restraints and/or protein-specific scoring functions. Many of those structural properties can be classified as one-dimensional structural properties because they can be represented as a one-dimensional vector along the protein sequence. This type of one-to-one prediction has been a computationally