PROTEIN DOMAIN BOUNDARY PREDICTION
The accurate delineation of protein domain boundaries is an important step for the prediction of protein structure, function, evolution, and design. Because a single domain spans an entire polypeptide chain or a subunit of such a chain, domains provide one of the most useful sources of information for understanding protein function, analysis based on domain families, and the study of individual proteins [1, 2].
Proteins comprise smaller building blocks, which are called “domains” or “modules.” These building blocks are distinct regions in a three-dimensional (3D) structure resulting in protein architectures assembled from modular segments that have evolved independently . The modular nature of proteins has many advantages, offering new cooperative functions and enhanced stability. For example, new proteins, such as chimeric proteins, can be created because they consist of multifunctional domains . The search method for templates used in comparative modeling can be optimized by delineating domain boundaries because the templates are classified based on domains . Domain boundary prediction can improve the performance of threading methods by enhancing their signal-to-noise ratio  and, for homologous domains, plays a key role in reliable multiple sequence alignment .
During the past three decades, numerous methods using the 3D coordinates of a protein structure ...