Study On The Prediction Of Protein Secondary Structure By Using Chemical Shifts

Protein is main structure molecular in building life and exercises the life activities. It is an important part of organism structure and function. The function of proteins is determined by its structure. To understand and master the function of proteins must first analyze the structure of the protein. How to predict protein structure is a very important research topic. The theoretical prediction is an important way to get protein spatial structure. Protein secondary structure as the basis of its spatial structure, studying on protein secondary structure is the theoretical foundation to understand protein of folding patterns and the tertiary structure. And it provides structure foundation to study the function of the protein and the interaction patterns between them. At the same time it can also help for new drug research and development. So it is necessary to study protein secondary structure prediction.In this study, we select 62 and 100 proteins with sequence identity of less than 30%. We used six nuclei chemical shifts (CSs) as features combined with the method of quadratic discriminant analysis (for short QD) to predict protein secondary structure, and the three-fold cross-validated test achieved better results. So, feature selection combining with algorithm on two data sets can effectively predict protein secondary structure. Moreover, we performed respectively the prediction by combining chemical shifts of six nuclei as features. The results illustrate that the chemical shifts of six nuclei play different role in the prediction of protein secondary structures.