Algorithmic and Artificial Intelligence Methods for Protein Bioinformatics

Chapter 17

Protein Tertiary Model Assessment

ANJUM CHIDA, ROBERT W. HARRISON, and YAN-QING ZHANG

17.1 Introduction

Protein structure prediction has been an important conundrum in the field of bioinformatics and theoretical chemistry because of its importance in medicine, drug design, biotechnology, and other areas. In structure biology, protein structures are often determined by techniques such as X-ray crystallography, NMR spectroscopy, and electron microscopy. A repository of these experimentally determined structures is organized as a centralized, proprietary databank called the Protein Data Bank (PDB). This databank is freely accessible on the Internet [1]. The generation of a protein sequence is much easier than the determination of protein structure. The structure of the protein gives much more insight about its function than its sequence. Therefore computational methods for the prediction of protein structure from its sequence have been developed. Ab initio prediction methods employ only the sequence of the protein based on the physical principles governing any molecular structure. Threading and homology modeling methods can build a 3D model for a protein of unknown structure from experimental structures of evolutionary related proteins. As long as a detailed physicochemical description of protein folding principles does not exist, structure prediction is the only method available to researchers to view the structure of some proteins. Experts agree that it is possible to ...