The Classified Model for DNA Sequences
|Course||Operational Research and Cybernetics|
|Keywords||DNA sequence codon frequency discriminating classincation Cluster|
According to nature of polarity of the forked chain of amino acids, it divides the base triplets into five categories, that is four kinds of amino acids and stop signal. By the appearance frequency5kinds of codon, we extract Characteristic Vector for representing DNA sequence. Base on the different content of different amino acids, the Characteristic Vector discloses the information of amino acids from two aspects of Content and arrangement of nucleotide bases.DNA sequence fragment is classified by Statistical techniques theory. DNA sequence fragment is classified by discriminating classification theory of Mahalanobis distance and Fisher discriminant method. The results showed that the positive rate of verified sample was100%and the consistent rate was90%. DNA sequence fragment is classified by Cluster theory and the positive rate of verified sample was95%,The results show that it is simple to the arithmetic and precision of classification results for the using of the biology knowledge and Lower dimension Characteristic vector. It This method is superior to method of discriminating that only considering base content.