Protein Structure Prediction using Homology Modeling: Methods and Tools

Introduction

Knowledge of the protein structure is an important step towards rational structure based drug design and virtual screening of large molecular libraries. Sequence-structure deficit marks one of the critical problems in today’s scenario where high-throughput sequencing has resulted in large datasets of protein sequences, but their corresponding 3D structures still need to be determined. Currently there are about 2 million protein sequences in Swissprot and TrEMBL, among them about 30,000 proteins have had their structures solved experimentally. Although the rate of experimental structure determination will continue to increase, the number of newly discovered sequences grows much faster than the number of structures solved. Since experimental structures can only be determined for a small fraction of proteins, computational methods for protein structure modeling play an increasingly important role yielding models suitable for a wide spectrum of applications.

The term “homology modeling”, also called comparative modeling refers to modeling of 3D structure of a protein by exploiting structural information from the known configurations of similar proteins. The method is based on the fact that structural conformation of a protein is more highly conserved than its amino acid sequence, and that small or medium changes in sequence normally result in little variation in the 3D structure. The necessary condition for successful homology modeling is a sufficient similarity (>30%) between the protein sequences. The importance and applicability of homology modeling is steadily increasing with the increase in the number of known structures in the protein structure database. Homology models contain sufficient information about the spatial arrangement of important residues in the protein, enabling us to study the binding site and design/dock drugs suitable for binding to the molecule. Stable and reliable repositories have been developed to give access to these annotated and evaluated models.

With the advent of new modeling softwares and algorithms, homology modeling is rapidly becoming the first choice for obtaining the 3D structure coordinates as well as other valuable structural and functional insights. The recent advancements in homology modeling, particularly in detecting and aligning sequences with Template structures, loop and side-chain modeling, model validation and evaluation have largely contributed to the prediction of more accurate models of protein structure, which was not possible several years ago. Improvement in sequence search/analysis, scoring systems and tertiary structure prediction methods allows development of models robust, error free models with high statistical significance. Publicly available, easy to use modeling servers and structure validation tools have further assisted the development of this technique.