In the machine learning field, the technique known as ensemble learning aims at combining different base learners in order to increase the quality and the robustness of the predictions. Indeed, this approach has widely been applied to tackle, with success, real world problems from different domains, including computational biology. Nevertheless, despite the potential of this technique, ensembles that combine results from different kinds of algorithms, have been understudied in the context of gene regulatory network inference. In this paper we used a genetic algorithm and frequent itemset mining, to study and design effective ensembles, to reverse-engineer gene regulatory networks, from high-throughput data. The methods proposed here, were evaluated and compared to well-established single and ensemble methods, on real and synthetic datasets. Results demonstrate the efficiency and the robustness of these new methods, advocating for their use as gene regulatory network inference tools.