Multiple sequence alignment using the Hidden Markov Model trained by an improved quantum-behaved particle swarm optimization

Multiple sequence alignment (MSA) is an NP-complete and important problem in bioinformatics. For MSA, Hidden Markov Models (HMMs) are known to be powerful tools. However, the training of HMMs is computationally hard so that metaheuristic methods such as simulated annealing (SA), evolutionary algorithms (EAs) and particle swarm optimization (PSO), have been employed to tackle the training problem. In this paper, quantum-behaved particle swarm optimization (QPSO), a variant of PSO, is analyzed mathematically firstly, and then an improved version is proposed to train the HMMs for MSA. The proposed method, called diversity-maintained QPSO (DMQPO), is based on the analysis of QPSO and integrates a diversity control strategy into QPSO to enhance the global search ability of the particle swarm. To evaluate the performance of the proposed method, we use DMQPSO, QPSO and other algorithms to train the HMMs for MSA on three benchmark datasets. The experiment results show that the HMMs trained with DMQPSO and QPSO yield better alignments for the benchmark datasets than other most commonly used HMM training methods such as Baum–Welch and PSO.