CIRDES: an efficient genome-wide method for in vivo RNA-RNA interactome analysis.

Complex RNA-RNA interactions underlie fundamental biological processes. However, a large number of RNA-RNA interactions remain unknown. Most existing methods used to map RNA-RNA interactions are based on proximity ligation, but these strategies also capture a huge amount of intramolecular RNA secondary structures, making it almost impossible to detect most RNA-RNA interactions. To overcome this limitation, we developed an efficient, genome-wide method, Capture Interacting RNA and Deep Sequencing (CIRDES) for in vivo capturing of the RNA interactome. We designed multiple 20-nt CIRDES probes tiling the whole RNA sequence of interest. This strategy obtained high selectivity and low background noise proved by qRT-PCR data. CIRDES enriched target RNA and its interacting RNAs from cells crosslinked by formaldehyde in high efficiency. After hybridization and purification, the captured RNAs were converted to the cDNA library after a highly efficient ligation to a 3' end infrared-dye-conjugated RNA adapter based on adapter ligation library construction. Using CIRDES, we detected highly abundant known interacting RNA, as well as a large number of novel targets of U6 snRNA. The enrichment of U4 snRNA, which interacts with U6, confirmed the robustness of the identification of the RNA-RNA interaction by CIRDES. These results suggest that the CIRDES is an efficient strategy for genome-wide RNA-RNA interactome analysis.

Identification of Pyroptosis Gene Signature Related Molecular Pattern, Clinical Implication, and Tumor Immunity in Hepatocellular Carcinoma`.

BACKGROUND: Pyroptosis is a novel form of programmed cell death in cancers, which regulates tumor cell invasion, proliferation, and metastasis, thereby affecting the prognosis of cancer patients. However, the role of Pyroptosis-Related Genes (PGs) in Hepatocellular Carcinoma (HCC) remains unclear. METHODS: Somatic mutation, copy number variation, and expression of 41 PGs were assessed in HCC and normal liver from the TCGA dataset. The Least Absolute Shrinkage and Selection Operator (LASSO) was used to construct the prognostic model. K-M curves, ROC curves, nomograph, and univariate and multivariate Cox regression were conducted to evaluate the predictive value of PGs. Immune infiltration was analyzed by CIBERSOFT and ssGSEA algorithm. The expression of prognostic PGs was validated by qPCR. RESULTS: Significant mutation and copy number variation of PGs were found in HCC. These genes were involved in an inflammatory response. In addition, 9 out of 41 PGs were differentially expressed in HCC and found to correlate significantly with patient survival. Then, these signature genes were selected to build a prognosis model and were utilized to stratify HCC patients into high and low PGs-score groups. It showed that the high-PGs group had a worse prognosis. Univariate and multivariate Cox regression verified that PGs-score was an independent risk factor for HCC. By ROC curves and nomogram, we showed that PGs-score effectively predicted the 1-, 3-, and 5-year survival of HCC patients and correlated with AFP level and disease stage. Immune infiltration analysis further showed that tumor immunity correlated with the PGs-score, and the expression of immune checkpoint molecule was significantly enhanced in the high PGs group. The PGs-score was also validated in the external validation cohort (ICGC). Finally, the expression of 9 signature genes was validated in normal liver and HCC cell lines. CONCLUSION: This study elucidated the aberrant regulation of PGs in HCC, and those pyroptosisrelated genes may be applied as a prognostic factor of HCC.