A meta-heuristic optimization tool for simplified protein structure prediction

Meta-heuristic algorithms give a satisfactory solution of complex optimization problems in a reasonable time. They are among the most promising and successful optimization techniques. However, some problems are highly complex and require improved techniques. A careful analysis of the existing meta-heuristic algorithms and hybridization among them may facilitate the research in this direction. To test this hypothesis, the author of the thesis developed a computational tool using a few meta-heuristic algorithms where these algorithms can be analyzed in detail and possible hybridization among them can be created. As a case study, the tool is developed for simplified protein structure prediction. The proper working of the software is demonstrated by optimizing the two sets of standard benchmark sequences. Along with testing and analyzing meta-heuristic algorithms, the tool can be used for simplified protein structure prediction.