Algorithmic and Artificial Intelligence Methods for Protein Bioinformatics

Chapter 7

Protein Structural Boundary Prediction

GULSAH ALTUN

7.1 Introduction

At the present state of the art in protein bioinformatics, it is not yet possible to predict protein structure [43]. Proteins are among the major components of living organisms and are considered to be the working and structural molecules of cells. They are composed of building block units called amino acids [25, 34]. These amino acids dictate the structure of a protein [49].

Many machine learning approaches and new algorithms have been proposed to solve the protein structure prediction problem [5, 7, 13, 15, 31, 41]. Among the machine learning approaches, support vector machines (SVMs) have attracted a lot of attention because of their high prediction accuracy [1]. Since protein data consist of sequence and structural information, another widely used approach for modeling these structured data is to analyze them as graphs. In computer science, graph theory has been widely studied; however, more recently it has been applied to bioinformatics. In out research, we introduced new algorithms based on statistical methods, graph theory concepts, and machine learning for the protein structure prediction problem.

In this chapter, we describe a novel encoding scheme and a computational method using machine learning for prediction starting and ending points of secondary structure elements. Most computational methods have been developed with the goal to predict the secondary structure of every residue in a given ...