Skip to Main Content
In the post-genomic era, understanding the interactions of genes plays a vital role in the analysis of complex biological systems. Recently, we developed a causal model approach for learning gene regulatory networks from microarray data. The optimization process for this learning was implemented by using genetic algorithm (GA) as a search technique to find the best candidate over the space of possible networks. In this paper, we propose a genetic algorithm which is guided by exploiting certain characteristics of diversity and high level heuristics in order to generate good networks as quickly as possible. A comparison of this algorithm to the standard genetic algorithms implemented in our earlier work is also presented in this paper. The Guided GA (GGA) is tested on both synthetic and real-world microarray data. The novel approach of GGA shows superiority of the solutions, computational efficiency along with accuracy improvement compared to standard GA.