Identification of PIWI-interacting RNA modules by weighted correlation network analysis

PIWI-interacting RNAs (piRNAs) are the largest class of small noncoding RNAs in animal cells; they cooperate with PIWI proteins to safeguard the genome. Although several studies have established that piRNAs can be biomarkers of cancer, it is still difficult to elucidate the exact function of piRNAs. In addition, researchers need to further investigate the interplay between piRNAs and piRNA groups in tumorigenesis. To identify cancer-associated piRNA-modules, we performed a weighted gene coexpression network analysis (WGCNA) on piRNA expression data from 11 types of cancer. Thereafter, genes associated with hub piRNAs in modules were predicted, and functional analysis of these genes was used to interpret the relation between piRNA and cancer. The results indicated that these piRNA modules have significant associations with cancer. A module with a high correlation coefficient (cor: -0.83, p value: 1.86E-128) was found; it was especially relevant to head and neck squamous cell carcinoma. Moreover, we found that hub piRNAs in modules may contribute to metastasis. This finding advances the understanding of piRNA function and its association with human cancer.