Research Of Swarm Intelligent Algorithm And Application To Biological Sequence Alignment

First of all, this paper provides an overview of swarm intelligence algorithms, and studies the principles and features of swarm intelligence algorithms in-depth. On this basis, an improved intelligent optimization algorithm is proposed, and the introduction of crossover to the quantum particle swarm optimization algorithm contributes to maintenance of the diversity in the later period of the search which makes the algorithm has better optimization results. After the optimization to the standard test functions, the experimental results show that the optimization effect of the new algorithm is significantly improved over the original algorithm that improves the effectiveness of the algorithm.Secondly, the multiple sequence alignment problems in the bioinformatics is studied, which is a NP complete problem. In this paper, the concept and key knowledge of alignment has made the introduction.The analysis of hidden Markov model based on the principle of multiple sequence alignment and the introduction of the three basic questions of HMM are also provided.Because of the easy falling into local optimal solution algorithm and the poor capabilities in global search problems, this paper combines intelligent algorithm QPSO with strong global search to a group with BW with fast local search algorithm for HMM training, whose model will be the base on the further multiple sequence alignment. The new algorithm have both the QPSO's global search ability and the Baum-Welch algorithm's fast approaching capacity of local optima .By the comparison of results between the original algorithm and the proposed alignment algorithm with test sequence from the standard library BAliBASE, we can see that the new alignment algorithm in this paper is effective. All in all, the No Free Lunch shows that we can't optimize the execution speed and the precision at the same time with a single optimization algorithm, so considering the advantages of a variety of optimization algorithms and making them work together can do better.