Hemoglobin protein secondary structure prediction by using improved clonal selection algorithm

The protein secondary structure information provides very important advantages in determining the function of a protein, treating numerous diseases and drug design. Determining the secondary structure in the laboratory environment is both costly and challenging. Therefore, the prediction of protein secondary structure has been an important study field of bioinformatics and computational biology for many years. The aim of the present two-phased study is to provide a contribution to the prediction of protein secondary structure using the nature-inspired methods. The data in the first phase were trained with Clonal Selection Algorithm (CSA) which was modeled by being inspired by the live immune system. CSA was improved by using the total similarity for the selection instead of the traditional way of the most similar antibody for an antigen. Using different methods classification is realized in the second phase. The classification was then performed with Multilayer Perceptron (MLP) which is modeled by being inspired by the biological nervous system, k nearest neighbor (kNN) algorithm which is based on distance, decision tree based ID3 method and Naive Bayes (NB) method based on statistical calculations. The results obtained indicated that the proposed antibody selection method performed better than the traditional method. The hemoglobin data set was trained by the new CSA model and the secondary structure prediction of the trained set was realized by different classification methods. In conclusion it was found out that the classification success of the trained data set was better than the success of the untrained data set.