|
|
Memetic algorithm for ab initio protein structure prediction in low resolution model |
|
|
Memetic algorithms (MAs) are a powerful combination of local search (LS) with standard evolutionary algorithms such as genetic algorithm (GA). Recent studies in various domains have shown that Memetic algorithms (MAs) are capable of effectively solving complex optimisation problems. MA also has the potential of choosing different local searches at different stages of evolutions leading to its adaptive application. The aim of the proposed research is to investigate the implementation of MA for the complex protein structure prediction (PSP) problem using the ab initio approach. The ab initio structure prediction attempts to investigate the protein folding process from its amino acid sequence without any structural information. To reduce complexity, low resolution protein models such as the well known HP model have been extensively used by researchers for studying the protein structure prediction and for the investigation of the dynamics that govern the protein folding process. Our preliminary results show the superiority of the memetic algorithm for PSP problem. |
|
|
Md. Kamrul Islam received his B.Sc. degree in Computer Science and Engineering from Bangladesh University of Engineering and Technology (BUET), Bangladesh. Md. Kamrul Islam commenced his PhD research in 2008 at Gippsland School of Information Technology (GSIT) in Monash University. Before starting his PhD, Mr. Islam was employed in TM International Bangladesh Ltd. (AKTEL) as a Sr. System Analyst. His current research focuses on application of computational intelligence techniques for ab initio protein structure prediction. |
|
