Clustering Analysis Of P53Tumor Suppressor Gene Based On Fuzzy Equivalence Relation

As a very important tumor suppressor gene, p53gene has become one of the hotspots in biology field. In this paper, the p53gene mRNA and protein sequences are studied, and we select the characteristic indexes of sequences and establish proximity relation of sequences, then we analyze p53gene sequences based on fuzzy clustering analysis method. Specific works of this paper are as follows:In chapter one, the content of bioinformatics and related knowledge and research of p53are briefly introduced. Besides, we give the main work and innovative points of this paper.In chapter two, firstly according to the properties of fuzzy equivalence relation, we use the weighted hamming distance method to establish fuzzy proximity relation of sequences, and then make fuzzy equivalence matrix to cluster the sequences. Then we select1.8human p53and its family members p63and p73tumor protein mRNA sequences. We take the base contents as index to make clustering analysis of sequences. Finally we analyze the difenences of the structures and functions of p53, p63and p73.In chapter three, firstly we divide twenty amino acids into four kinds based on the detailed HP model. Then we establish a continuous coordinate space, and use feature vector to describe protein sequences through Chaos Game Representation method, and define the distance of the sequences. Finally we make similarity analysis and fuzzy clustering of p53gene protein sequences of eight species including human.In chapter four,18kinds of p53protein sequences are studied and the amino-acid future composition of p53protein sequences is worked out through the amino-acid mutating probability. Then we establish four-dimensional vector to depict p53protein sequences based on the future contents of non-polar amino-acid, negative amino-acid, non-charged polar amino-acid and positive amino-acid of protein sequences. We calculate the similarity of sequences using the cosine of the angle of the vector and establish the fuzzy proximity matrix of sequences. And we cluster the sequences based on the cut relation of the transitive closure of the fuzzy proximity matrix. Finally, we test the clustering result by F-test and determine the optimal number of clustering.In chapter five, we sum up the content of the paper and give prospect of the further study of p53gene sequences.