Hybrid High Exploration Particle Swarm Optimization Algorithm Improves the Prediction of the 2-Dimensional Hydrophobic-Polar Model for Protein Folding

Author(s): Cheng-Hong Yang, Yu-Shiun Lin, Sin-Hua Moi, Kuo-Chuan Wu, Li-Yeh Chuang*, Hsueh-Wei Chang*

Journal Name: Current Bioinformatics

Volume 13 , Issue 2 , 2018

Become EABM
Become Reviewer
Call for Editor

Graphical Abstract:


Background: Protein folding depends on the nature of the amino acid sequence. Once folding process of the amino acid sequence is successful, the protein becomes functional. Recently, a two-dimensional hydrophobic-polar (2D HP) model algorithm has been developed for the effective prediction of protein folding. However, the particular 2D HP models still lack an algorithm for protein folding prediction. Objective: Some developed algorithms still require further improvement in terms of accuracy and search stability.

Method: In order to evaluate its improvement for protein folding of the 2D HP model in this study, we propose the hybrid high exploration particle swarm optimization (HHEPSO) method, which employs the HEPSO algorithm for optimization which combines both hill climbing and greedy algorithms for local search.

Results: Several algorithms for protein structure prediction on the 2D square and triangular lattice models are compared with HHEPSO. In terms of accuracy and stability, our proposed HHEPSO revealed better performance than most of the test algorithms. HHEPSO also successfully deals with protein structure prediction problems for the longer amino acid sequences.

Conclusion: Our proposed HHEPSO algorithm is accurate and effective for protein structure prediction for a 2D triangular lattice model.

Keywords: Hybrid algorithm, HHEPSO, particle swarm optimization, hill climbing algorithm, greedy algorithm, protein folding, hydrophobic-polar (HP) model.

Rights & PermissionsPrintExport Cite as

Article Details

Year: 2018
Published on: 26 March, 2018
Page: [182 - 192]
Pages: 11
DOI: 10.2174/1574893612666171006161527
Price: $65

Article Metrics

PDF: 15
PRC: 1