Multi-objective evolutionary strategy approaches for protein structure prediction
Shuangbao Song · Institutional Repositories DataBase (IRDB) · 2020
Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.
Institutional Repositories DataBase (IRDB) · National Institute of Informatics · dissertation · 2020
Abstract
solvent accessible surface area.In AIMOES, an evolution scheme which flexibly reuse past search experiences is incorporated to enhance the efficiency of conformation search.A decision maker based on the hierarchical clustering is carried out to select representative solutions.A set of benchmark proteins with 30 ∼ 91 residues is tested to verify the performance of the proposed method.Experimental results show the effectiveness of AIMOES in terms of the root mean square deviation (RMSD) metric, the distribution diversity of the obtained Pareto front and the success rate of mutation operators.The superiority of AIMOES is demonstrated by the performance comparison with other five state-of-the-art PSP methods.This thesis is organized as follows: Chapter 1 gives a brief introduction about the PSP problem and multi-objective optimization.Chapter 2 presents some important concepts.Chapter 3 presents the energy function used in these two methods.In Chapter 4, we shows the multi-objective evolutionary strategy where solvent effect are incorporated into, i.e.MO3, for solving the PSP problem.The experimental results of MO3 are also shown in this chapter.Then, in Chapter 5, the archive information assisted multi-objective evolutionary strategy, i.e.AIMOES, for solving the PSP problem is described.Finally, we draw the conclusions of this thesis in Chapter 6.
Open the resolved locator · 10.15099/00020249
Metadata licence: OpenAlex metadata (CC0)