The Construction And Application Of RNA-structure-based Elastic Network Model

RNA is a very important bio-macromolecule in the living organisms.For a long time,it is well known that RNA only acts as a messenger for the transfer of genetic information existed in the living organisms.However,with the developing of the molecular biology research,it is increasingly accepted that RNA also folds into the specific tertiary structure,similar to the protein,and also play important biological functions in many biochemical processes in the organisms,such as the translation and synthesis of proteins,the catalysis of chemical reactions,the regulatory of gene expressions,the growth and development of the organisms.The performance of RNA biological functions is relied on its specific three-dimensional structure.It is an unresolved scientific question how to predict the RNA functions from its tertiary structure.When RNA plays its function,it usually undergoes certain conformational motions,instead of standing stationarily.Therefore,besides the static tertiary structure of RNA,it is necessary to obtain the dynamic properties encoded in the RNA structures.It is an important problem,worthy of in-depth study,how to investigate the intrinsic dynamical properties of RNA from its three-dimensional structure,and then reveal the molecular mechanism for the performing of RNA functions.The Elastic network model(ENM)is an important tool to obtain the intrinsic dynamical properties encoded in the structure of biological macromolecules,which is widely used in the investigating of the structure-function relationship of proteins.However,for RNA,the application effect of ENM is poor,especially for the small RNA with loose structures.In order to solve this problem,some preliminary studies have shown that the application effect of ENM becomes better when constructing more complex models.We think that besides constructing the more complex ENM,it is also an efficient method that optimizing the spring coefficients of ENM to improve the prediction effects of this model.In this thesis,the ENM suitable for RNA structure is constructed,and then the spring coefficients in the ENM are optimized,in which the good results are obtained.In our studies,the RNA structure is simplified as an elastic network,where each nucleotide acid is simplified as three nodes which represent the ribose,base and phosphate groups,respectively.Then,the interactions between these nodes are represented by springs.By this way,the ENM suitable for RNA is constructed.Besides that,in our model,the interactions between nodes are classified into covalent and non-covalent interactions,and then the spring coefficients for these two types of interactions are distinguished.The values for these two spring coefficients are optimized by using the experimental data of the temperature factors to improve the prediction effects of ENM in the study of the RNA structure-function relationship.