Biclustering Analysis of Transcriptomic Big Data Identifies Condition-specific microRNA targets

Abstract

We present a novel approach to identify human microRNA (miRNA) targets for a variety of cell conditions by biclustering a large collection of mRNA fold-change data for sequencespecific targets. The bicluster targets exhibited on average 17.0% (median 19.4%) improved gain in certainty (sensitivity + specificity). The net gain was further increased up to 32.0% (median 33.2%) by incorporating functional networks of targets. We analyzed cancer-related biclusters and found that PI3K/Akt signaling pathway is strongly enriched in targets of a few miRNAs in breast cancer and diffuse large B-cell lymphoma. Among them, five independent prognostic miRNAs were identified, and repressions of bicluster targets and pathway activity by mir-29c were experimentally validated. In total, 29,898 biclusters for 459 human miRNAs were collected in BiMIR database, where biclusters are searchable for miRNAs, tissues, diseases, keywords, and target genes.