The Protein Structure Classification Research Based On Data Mining

Protein structure classification is based on several levels. In recent years, the topic is to find methods to classify and analyze automatically protein structure. Protein second structure data is chosen as main study object, and data mining and dynamic programming are applied to protein structure classification. It is introduced that a new protein structure classification method: PSSC. STRIDE and ISSA are primary algorithms in the paper. A classification model is a main part of PSSC that includes establishing CATH classification tree, calculating protein structure similarity, setting classification standard based statistics. In the classification of protein structure with four levels, RMBS, Z-Score and ISSA algorithm are applied to obtain parameters. Lastly, scientific computing software is introduced for PSSC. Besides those, something needs to be improved, which are described and given solutions in the paper.