An integer programming model for predicting multi-shapes of 3D protein structure model

Abstract

Protein structure prediction is using bioinformatics and computational science tools is a vital part of medication discovery and infection expectation, but calculating the structure of a folded protein is a difficult task. In this paper, we predict protein structure of hydrophobic-polar lattice models in three dimensions. The considered problem is treated as an integer programming problem. The model finds multiple optimal shapes with the same minimal energy of the protein. Optimization is performed by Tabu search and the sequences are investigated by blending two fundamental techniques: strengthening and enhancement. The effectiveness of our strategy was confirmed in comparison with existing approaches benchmark protein sequences.