MOGARS: Multi-objective genetic algorithm for the discovery of regulatory sequences

This paper proposes the use of a multi-objective approach to discover regulatory sequences using genetic algorithms. Since regulatory sequences have different features, a multi-objective approach seems pertinent and a viable way. The variables to maximize with it are: similarity, length, support, and a weight matrix intended to achieve a guided search. We validated our proposal in a database of the Bacillus Subtilis bacteria. Our experimental results were considered as valuable discoveries by domain experts.