A Particle Swarm Optimization-Based Approach with Local Search for Predicting Protein Folding

Area of Science:

• Computational biology
• Biophysics
• Bioinformatics

Background:

• The hydrophobic-polar (HP) model is a fundamental tool for understanding protein folding and hydrophobic interactions.
• Accurate prediction of protein structures is crucial for understanding biological function and disease.

Purpose of the Study:

• To develop and evaluate a novel algorithm for predicting protein folding structures within the HP model.
• To enhance the efficiency and accuracy of protein structure prediction using metaheuristic optimization.

Main Methods:

• Development of the high exploration particle swarm optimization with local search (HE-L-PSO) algorithm.
• Integration of global search (HEPSO) with local search techniques (hill-climbing, greedy, Tabu table).
• Validation of HE-L-PSO using 20 known protein structures of varying lengths.

Main Results:

• HE-L-PSO achieved optimal solutions for all tested protein folding structures.
• The algorithm demonstrated high reproducibility and stability in predictions.
• HE-L-PSO outperformed seven other reported algorithms in predicting both short and long amino acid sequences.
• Predicted structures exhibited a hydrophobic core consistent with native protein folding.

Conclusions:

• The HE-L-PSO algorithm is a highly effective tool for protein structure prediction in the HP model.
• This approach offers improved accuracy, stability, and reproducibility compared to existing methods.
• The findings contribute to advancing computational methods in structural biology and protein design.