which generates a bias in the development and application of biclustering. This bias has two main consequences:first, gene expression datasets act as a benchmark both during the development of new biclustering algorithms and in their independent comparison, meaning that these new ...