Study On Spatially Long-ranged Correlation Within Lysozyme

Study on Spatially Long-ranged Correlation within Lysozyme Proteins are essential parts of organisms and participate in virtually every process within cells.Protein structure and function has been one of the important issues in science because of the importance of protein.It is extremely important to study protein spatial configuration for explaining protein structure and function,and dynamic information of protein conformational changes is of great value.Protein allostery requires dynamical structural correlations,and most biologically relevant allostery are spatially long-ranged(SLR).Therefore,understanding of structural correlations,especially SLR ones,are essential for elucidation and manipulation of protein allostery.Based on analysis of molecular dynamics(MD)simulation trajectories for lysozyme,we found stronger backbone dihedral angles pairs correlation even though the pairs are spatially long-ranged,and predicted possible functional sites for protein allostery.In this thesis,a calculation method of spatially long-ranged correlation is proposed.First of all,we analyzed the distribution of every backbone dihedral angle of HEWL.Because protein structure is very stable,the vast majority of the dihedral angles have only one state,but some have two or more which may be key sites for allostery.Calculating circular correlation coefficient(CCC),mutual information(MI),interaction information(II)and distance of backbone dihedral angles,we analyzed long-ranged correlation,and found that mutual information is better to analyze correlations.We also analyzed the relationship between MI and structure,and which between MI and distance,and found that spatially long-ranged correlations are dominantly executed by loop residues.Next,based on the difference between two states of a dihedral angle,all snapshots in the trajectory were divided into two subdatasets,then the difference of MI between two subdatasets was calculated.We found that the correlation between two backbone dihedral angles can be influenced by other ones which may be far away in space.We analyzed the relationship between the difference of MI and distance to clarify what we found.Then,in order to explain that the correlation between two dihedral angles can be influenced by other ones which may be far away in space,we calculated the interaction information of every three backbone dihedral angles,and found that the long-ranged correlation exists even when the angle is spatially far away from the others and this method is more reliable to analyze the correlation of every three.What's more,we calculated MI between every two residues like backbone dihedral angles,the result also well reflected obvious correlation between remote residues,which indicated that the method using MI to analyze correlation is reliable and can be applied to analyze correlations between residues groups or domains.Finally,based on the analysis of long-ranged correlation of backbone dihedral angles and residues,we predicted some possible key sites for allostery and compared these sites with the known sites.The analysis of long-ranged correlation can help to build correlation network model of protein,and implicate a general search strategy for novel allosteric modulation sites of protein activities.