An Advanced Twin Support Vector Machine Algorithm for Gene Splicing Sites Prediction

Abstract Gene splicing sites prediction is of great significance, not only for the large-scale and high-quality computational annotation of genomes, but also for the analysis and prediction of the gene sequences evolutionary process. It is helpful to promote the development of smart healthcare. Traditional Twin Support Vector Machines (TWSVM) algorithm has advantages in solving small-sample, nonlinear and high-dimensional problems, but it cannot deal with parameter selection well. Therefore, a Particle Swarm Optimization Twin Support Vector Machines (PSO-TWSVM) algorithm for gene splicing sites prediction was proposed in this paper. In order to avoid the blindness of parameter selection, particle swarm optimization algorithm was used to find the optimal parameters of twin support vector machines. The proposed algorithm was compared with traditional TWSVM algorithm and Least Squares Support Vector Machine (LSSVM) algorithm. The comparison results show that the positive sample prediction rate, negative sample prediction rate and correlation coefficient (CC) of the proposed algorithm are the best among the three different support vector machines. The proposed algorithm effectively improves the prediction rate and the accuracy of splice sites. The comparison experiments verify the feasibility of the proposed algorithm.