| Summary: | 碩士 === 國立中山大學 === 資訊工程學系研究所 === 92 === It is well known that the biological function of a protein depends on its 3D structure. Therefore, solving the problem of protein structures is one of the most important works for studying proteins. However, protein structure prediction is a very challenging task because there is still no clear feature about how a protein folds to its 3D structure yet.
In this thesis, we propose a genetic algorithm (GA) based on the lattice model to predict the 3D structure of an unknown protein, target protein, whose primary sequence and secondary structure elements (SSEs) are assumed known.
Hydrophobic-hydrophilic model (HP model) is one of the most simplified and popular protein folding models. These models consider the hydrophobic-hydrophobic interactions of protein structures, but the results of prediction are still not encouraged enough. Therefore, we suggest that some other features should be considered, such as SSEs, charges, and disulfide bonds. That is, the fitness function of GA in our method considers not only how many hydrophobic-hydrophobic pairs there are, but also what kind of SSEs these amino acids belong to. The lattice model is in fact used to help us get a rough folding of the target protein, since we have no idea how they fold at the very beginning.
We show that these additional features do improve the prediction accuracy by comparing our prediction results with their real structures with RMSD.
|
|---|
