New multiple sequence alignment approach reveals proteins structural determinants.

摘要

In today's post genome era, protein analysis including the ability to extract knowledge from completed genome projects has become the focus of many research projects. One area of protein analyses that gained significant attention is the ability to determine structural determinant residues known for their significant role in polypeptide folding and protein structure stability. These residues are found at various locations such as the protein surface area and within important structural regions such as the hydrophobic core or the nucleus of protein fold. Knowledge of structural determinant residues is important for many research areas including protein structural analysis, classification, sequence-structure relationships and the design of pharmaceutical drugs. Today, using experimental or computational methods, discovery of protein structural determinants remains a slow and challenging endeavor. While the experimental methods are complex and consume time and resources, computational methods did not improve discovery of structural determinant as many of these methods are still evolving or have known limitations. Additionally, most structural determinant research primarily focus on finding residues localized to small protein regions such as those in on the surface area or within the hydrophobic core. Thus, these challenges and others create an urgent need to develop and implement new methods to enhance the ability to detect unknown structural determinants for different groups of proteins.;The goal of this investigation was to determine key and conserved amino acid residues that potentially are the structural determinant for groups of so called beta sandwich-like proteins. This goal was aligned with the project objectives to develop and implement a novel multiple protein alignment approach to enable detection of a wide range of structural determinant residues located at diverse regions of a set of non-similar proteins. This set included the protein domains from diverse families and superfamilies where the level of amino acids residues similarity is low while these proteins shared similar arrangement of secondary structural elements. Furthermore, the achievement of these project goals required reliance on the protein classification scheme based on structural similarities between the compared proteins that supported the alignment of protein domains independent of amino acids residues similarities.