Investigation of the three-dimensional lattice HP protein folding model using a genetic algorithm

Fábio L Custódio,Hélio J C Barbosa,Laurent E Dardenne

doi:10.1590/s1415-47572004000400023

Fábio L Custódio, Hélio J C Barbosa + Show 1 more

Open Access

https://doi.org/10.1590/s1415-47572004000400023

Copy DOI

Abstract

An approach to the hydrophobic-polar (HP) protein folding model was developed using a genetic algorithm (GA) to find the optimal structures on a 3D cubic lattice. A modification was introduced to the scoring system of the original model to improve the model's capacity to generate more natural-like structures. The modification was based on the assumption that it may be preferable for a hydrophobic monomer to have a polar neighbor than to be in direct contact with the polar solvent. The compactness and the segregation criteria were used to compare structures created by the original HP model and by the modified one. An islands' algorithm, a new selection scheme and multiple-points crossover were used to improve the performance of the algorithm. Ten sequences, seven with length 27 and three with length 64 were analyzed. Our results suggest that the modified model has a greater tendency to form globular structures. This might be preferable, since the original HP model does not take into account the positioning of long polar segments. The algorithm was implemented in the form of a program with a graphical user interface that might have a didactical potential in the study of GA and on the understanding of hydrophobic core formation.

Highlights

Determining the functional conformation of a protein molecule from amino acid sequence remains a central problem in computational biology (Pedersen and Moult, 1996)
The hydrophobic-hydrophilic model (Dill and Lau, 1989) describes the proteins based on the fact that hydrophobic amino acids tend to be less exposed to the aqueous solvent than the polar ones resulting in the formation of a hydrophobic core in the spatial structure
This work is an investigation of the HP model in a three-dimensional cubic lattice using a genetic algorithm (GA) as a tool to find the optimal conformation for a given sequence

Summary

Introduction

Determining the functional conformation of a protein molecule from amino acid sequence remains a central problem in computational biology (Pedersen and Moult, 1996). Since the free energy on the original model is given only by the number of non-specific hydrophobic contacts, it is a fact that the positions of polar segments are not directly optimized when searching for optimal structures. This work is an investigation of the HP (hydrophobic-polar) model in a three-dimensional cubic lattice using a genetic algorithm (GA) as a tool to find the optimal conformation for a given sequence.

Results

Conclusion