Improved Hidden Markov Model training for multiple sequence alignment by a particle swarm optimization—evolutionary algorithm hybrid

Biosystems ◽  
2003 ◽  
Vol 72 (1-2) ◽  
pp. 5-17 ◽  
Author(s):  
Thomas Kiel Rasmussen ◽  
Thiemo Krink
Keyword(s):  
Particle Swarm Optimization ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Evolutionary Algorithm ◽  
Sequence Alignment ◽  
Multiple Sequence Alignment ◽  
Hidden Markov ◽  
Multiple Sequence ◽  
Swarm Optimization ◽  
Model Training

Multiple Sequence Alignment Based on Profile Hidden Markov Model and Quantum-Behaved Particle Swarm Optimization with Selection Method

2011 ◽  
Vol 282-283 ◽  
pp. 7-12 ◽  
Author(s):  
Hai Xia Long ◽  
Li Hua Wu ◽  
Yu Zhang
Keyword(s):  
Particle Swarm Optimization ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Sequence Alignment ◽  
Multiple Sequence Alignment ◽  
Hidden Markov ◽  
Particle Swarm ◽  
Multiple Sequence ◽  
Swarm Optimization ◽  
Profile Hmm

Multiple sequence alignment (MSA) is an NP-complete and important problem in bioinformatics. Currently, profile hidden Markov model (HMM) is widely used for multiple sequence alignment. In this paper, Quantum-behaved Particle Swarm Optimization with selection operation (SQPSO) is presented, which is used to train profile HMM. Furthermore, an integration algorithm based on the profile HMM and SQPSO for the MSA is constructed. The approach is examined by using multiple nucleotides and protein sequences and compared with other algorithms. The results of the comparisons show that the HMM trained with SQPSO and QPSO yield better alignments than other most commonly used HMM training methods such as Baum–Welch and PSO.

Sign in / Sign up

Export Citation Format

Share Document