Establishing A Database Of Homo Pre-mRNA Secondary Structures And Inquiring Into The Correlation With Proteins

The research on RNA secondary structure is a very important issue in bioinformatics, and it has received a lot of development and deepening. In contrast to the RNA primary structure, the diversity of its secondary and high-class structure possesses rich biological significance. Therefore, the RNA structure is the key to probe into RNA complicated functions and characteristics. Because of the difficulties in?purification, crystallization and decomposition of RNA, it's not easy for experiment methods such as X-ray crystallography and NMR to determine RNA 3D-structures. Moreover, it would take up lots of time and energy. Therefore, we should use the theory to predict RNA structures with the help of mathematics and computer technology, and get the relationship between the structure and the function of RNA.Early in 1981, Zuker proposed the free energy minimization algorithm. With nearly 30 years improvements, this algorithm has been widely used to predict RNA secondary structure. In this paper, we predicted a large number of pre-mRNA secondary structure of human samples by using Zuker's free energy minimization algorithm and the dynamic extended folding simulation that we have developed, and stored the results as a database. Then we analyzed the relation between the secondary structures of pre-mRNA and protein.The thesis mainly includes three parts of following contents and conclusions:1. The pre-mRNA secondary structure database of human samples have been built. We collected 335 pre-mRNAs of human samples from GenBank database, and authored a program according to the dynamic extended folding simulation, so that the most steps of dynamic extended folding simulation have been programmed completely. As the foundation of the successor, we built the pre-mRNA secondary structure database of human samples, and released it on the server of Yunnan University (http://www.mbrc.ynu.edu.cn/Human_mRna_db.html) for all scientists.2. We built a database on the relation of mRNA and its coding protein. We searched the coding protein information of 335 human pre-mRNA samples in the database, and achieved the matching of the sequence of mRNA, the secondary structure of mRNA, the sequence of amino acid and the secondary structure of protein by program. Finally, we saved the results as a specified format for the following discussion.3. From the above results, the following conclusions were drawn. The mRNA secondary structures corresponding to those regular secondary structures in proteins usually have a higher value of m/n, accordingly, the mRNA secondary structures corresponding to the irregular have a lower value of m/n. The paired regions in the mRNA secondary structures have a higher value of m/n, the unpaired regions low. The paired regions of mRNA tend to code the regular secondary structures in protein, while the unpaired regions tend to code the irregular secondary structures in protein.