Identifying human miRNA targets with a genetic algorithm

MicroRNAs (miRNAs) play an important role in eukaryotic gene regulation. Although thousands of miRNAs have been identified in laboratories around the world, most of their targets still remain unknown. Different computational techniques exist to predict miRNA targets. In this article, we propose a new method for identifying human miRNA-mRNA interactions based on a genetic algorithm. Our cross-validation results indicate that the genetic algorithm-based miRNA target predictor outperforms the MiRanda package as evidenced by high true positive rates and moderate false positive rates.