Method for the identification of the subsets of genes specifically consistently co-expressed in a set of datasets

The recently proposed binarization of consensus partition matrices (Bi-CoPaM) ensemble clustering method has offered the ability to mine multiple genome-wide microarray datasets for the subsets of genes which are consistently co-expressed in all of these datasets. Though, some of those subsets of genes might also be consistently co-expressed in many other datasets that were generated under a wider range of conditions than those of interest in a single focused study. Here we propose a new method, named as the unification of clustering results from multiple datasets using external specifications (UNCLES). The external specifications imposed in this study aim at mining for the subsets of genes that are consistently co-expressed in one set of datasets (S⁺) and not consistently co-expressed in another set of datasets (S^-). We tested our proposed method over eight budding yeast cell-cycle datasets for S⁺ and other six general budding yeast datasets for S^-. Our results have shown the ability of our method to find the subsets of genes consistently co-expressed in the S⁺ datasets successfully, while excluding the subsets of genes that are also consistently co-expressed in the S^- datasets.