An integrative framework for identifying consistent microRNA expression signatures associated with clear cell renal cell carcinoma

Clear cell renal cell carcinoma (ccRCC) is the most common and invasive renal-originated malignancy. Altered microRNA expression has been observed in many human cancers including ccRCC. Microarray is routinely used in labs worldwide for detecting cancer specific microRNA expression profiles, but no consistent conclusion could be drawn so far. The function of microRNAs in carcinogenesis of this tumor type is thereof largely unknown. In this study, we describe an integrative framework to improve the comparability of differentially expressed microRNAs (DE-miRNAs) from different experiments, and apply it to 4 publicly available microRNA expression datasets in ccRCC. The approach uses a novel statistic method for cancer outlier detection. The identified DE-miRNAs are then screened by POMA, an in-house developed predictor, for microRNAs with real regulatory activity in the disease. The proposed framework not only achieves high reproducibility across different datasets but also identifies a consistent set of 12 DE-miRNAs which could be putative biomarkers and therapeutic targets. The targets of DE-miRNAs in each dataset were then mapped to functional databases for enrichment analysis. Both novel and previously characterized microRNA-regulated molecular pathways are identified that are likely to contribute to the pathogenesis of ccRCC. Overlapping comparison suggests that independent ccRCC expression profiles are more consistent at pathway level than that at gene/microRNA level.