Fig. 1

Pipeline of the miRspongeR package. The pipeline mainly contains three components: Identification of miRNA sponge interactions, Identification of miRNA sponge modules, and Validation and analysis. For the identification of miRNA sponge interactions, out of the eight implemented methods, seven (miRHomology, pc, sppc, ppc, hermes, muTaME and cernia) are stand-alone methods and one (integrateMethod) is an ensemble method which integrates the prediction results of the component methods. To understand the module-level properties of miRNA sponges, four module identification methods (FN, MCL, LINKCOMM and MCODE) are provided to identify miRNA sponge modules based on the identified miRNA sponge interaction networks. For the validation of miRNA sponge interactions, the ground truth data is used to validate predicted miRNA sponge interactions. Furthermore, enrichment analysis is performed to identify potential diseases, biological processes and pathways associated with miRNA sponge modules. Survival analysis is also performed to identify significant miRNA sponge modules which can distinguish the high and low risk tumor samples. Users can prepare their own datasets as the input to miRspongeR