The Analysis Of Protein Spatial Structure Similarity Comparison

Always, similarity comparison of protein structures plays an important role in the computational molecular biology. Predicting the structure and function of protein has been becoming one of the hottest points in present life science. The21st century is the post-genomic era. More and more sciences try their best to find the relationship between the function and the three-dimensional structure. Traditional methods that be used to analysis those one-dimensional amino acid sequences usually have obvious advantage in the time and space complexity. But they also have some limitations. Ignoring the three-dimensional structure will leads to lost of the biological information in some degree. Therefore, one-dimensional methods can not take place with the three-dimensional methods. Inevitably, the similarity comparison of protein structures based on one-dimensional methods will be accepted by more and more people in the future.In this dissertation, first, we simply review several existing methods. Secondly, we summarize those old methods and then propose two new methods to compute the similarity between two different proteins. The first method bases on the frequency matrix. We use the frequency matrix to establish a fuzzy matrix, and then compute the equivalent matrix, base on it we can get the final result. In the second method, we make use of a cube model in the three-dimensional space. Then we base on some rules to transform sequences to a series of points in the space, after it, we can get some fitting curves through curve fitting. Also we can get the distribution of those points, and then make a intuitive conclusion. The next we pick up the characteristic value of fitting curves, using these values we can calculate the accurate similarity between two different proteins.Finally, we make analysis for some species sequences using the novel methods. Because of the complexity in count, we work out a C++program to solve it.