Recent studies have discovered that miRNA sponges are closely associated with the occurrence and development of cancers, and it is a significant and challenging subject to study the mechanisms of miRNA sponges in cancers. From the perspective of systems biology, this subject integrates heterogeneous biological data including high throughput data, such as miRNA, lncRNA, mRNA, pseudogene and circRNA expression profiles, miRNA-target databases, and genes-disease relationship databases, to study the identification and reconfirmation of cancer miRNA sponges, the causal relationships between cancer miRNA sponges, and new methods and technologies of miRNA, lncRNA, mRNA, pseudogene and circRNA expression data analysis. The aim is to identify cancer-related miRNA sponges and miRNA sponge causal networks and modules in cancers, and further develop a bioinformatics analysis software package for inferring miRNA sponge causal networks and modules in cancers. By combining molecular function databases (GO, KEGG, etc.), we apply the methods into Pan-Cancer data analysis including 12 disease types to mine cancer-conserved and specific miRNA sponge causal networks and modules. The research results will be of great importance in exploring the pathogenesis mechanisms of cancers, and promoting the study on precision medicine of human cancers.