MiRComb: An R Package to Analyse miRNA-mRNA Interactions. Examples across Five Digestive Cancers
Fig 2
Main findings of the LIHC report.
A) Principal Components Analysis (PCA) (based on correlation matrix) of miRNA samples. B) Volcano plot showing the miRNAs according to its logratio between cancer and control. C) Heatmap of the top 50 most deregulated miRNAs according to its FDR. D) Density plot of the Pearson Correlation Coefficients of all possible miRNA-mRNA interactions. Lines show different cutoff: p-value < 0.05, p-value < 0.01, FDR < 0.05 and FDR < 0.01. E) Correlation of miR-139-5p and CCNB1 as an example. F) Venn diagram showing the total number of sigifnicant correlations (FDR < 0.05), the total number of predicted interactions in at least one database (TargetScan or microcosm), and the intersection of both. G) Network of selected interactions. Each miRNA-mRNA interaction is negatively correlated (FDR < 10–33) and predicted at least in one database (Targetscan or MicroCosm). Circles represent miRNAs and squares mRNAs; red fill means upregulated miRNA/mRNA, while green fill means downregulated miRNA/mRNA; lines indicate the miRNA-mRNA pairs; red line means positive score and green line means negative score; arrow width is proportional to the number of appearances on the databases (TargetScan or MicroCosm). H) Pie chart showing the number of mRNAs regulated by 0, 1, 2, 3, 4, 5, and >5 miRNAs. I) Barplot showing the number of targets per miRNA and the percentage of mRNAs that are cumulatively regulated by the miRNAs. J) Circos plot of the top 45 miRNA-mRNA interactions sorted by FDR, a line means a miRNA-mRNA pair. Blue lines are the position of the miRNAs and orange lines are the position of the mRNAs.