An Evolutionary Algorithm with Gene Ontology-Aware Crossover Operator for Protein Complex Detection

Evolutionary algorithms (EAs), as global search methods, are proved to be more robust than their counterpart local heuristics for detecting protein complexes in protein-protein interaction (PPI) networks. Typically, the source of robustness these EAs comes from components and parameters. These solution representation, selection, crossover, mutation. Unfortunately, almost all EA based complex detection methods suggested literature were designed with only canonical or traditional components. Further, topological structure network is main information that used design such The contribution this paper formulate a biological consistency. For purpose, new crossover operator where terms both gene semantic similarity functional fed into its design. To reflect heuristic roles similarities, introduces two ontology (GO) aware operators. direct annotation-aware inherited first strategy handled annotation proteins, while second directed acyclic graph (DAG) each term product. conduct our experiments, proposed GO-aware operators compared against state-of-the-art heuristic, operator, GO-based EAs. Simulation results evaluated recall, precision, F measure at level level. prove encourages reliable treatment exploration exploitation and, thus, improves ability accurate structures.