Composite Approaches to Protein Tertiary Structure Prediction: A Case‐Study by I‐Tasser

Abstract

Protein structure prediction methods have been historically categorized into comparative modeling, threading, and ab initio modeling. With rapid progress in the field, the borders between different categories have become increasingly blurred and the most efficient approaches appear to be those which use hybrid algorithms. In this chapter, we review the progress and development of composite approaches for protein structure prediction. As a case study of the composite approach, we introduce in detail the algorithm of I-TASSER, a fully automated and composite approach consisting of initial template identification, fragment reassembly, ab initio loop modeling, and atomic-level structure refinement. The strength and weakness of I-TASSER in benchmark tests and the blind CASP experiments are discussed. The I-TASSER server is freely available to the academic community at http://zhang.bioinformatics.ku.edu/I-TASSER.