Additional file 3 of Crosstalk between microRNA expression and DNA methylation drives the hormone-dependent phenotype of breast cancer

Additional file 3. a) Overview and annotation of the 119 mimQTL miRNAs. b) Overview and annotation of the 26,746 mimQTL CpGs. c) Genes found positively or negatively correlated to miRNAs of each cluster (mRNA-miRNA Spearman correlation > 0.4 or <-0.3, respectively, in both the Oslo2 and TCGA cohorts). d) Enrichr [33] Pathway enrichment (KEGG 2019 Human database) of genes positively (top) and negatively (bottom) correlated to miRNAs of each cluster. e) miRNA expression correlation to the Nanodissect [34] lymphocyte infiltration score, the xCell [35] fibroblast infiltration score, and ESR1 mRNA expression. f) Differential expression of miRNAs between clinically relevant breast cancer groups. g) Results from generalized linear modeling (GLM) of miRNA expression as a multivariate function of lymphocyte and fibroblast infiltration and ESR1 mRNA expression. h) Pathway enrichment using Enrichr [33] of genes mapped to the CpGs of cluster 1 according to the Illumina HumanMethylation450k array. i) ChromHMM [37] enrichment of genomic regions mapped to the CpGs of cluster 1 and 2. j) Pathway enrichment using Enrichr [33] of genes mapped to the CpGs of cluster 2 according to the Illumina HumanMethylation450k array. k) Table of 273 unique mimQTL associations with 69 unique CpGs residing in miRNA super-enhancer regions. l) Long-range loops overlapping with mimQTLs residing on the same chromosome. m) Correlation between the Global Methylation Alteration (GMA) score and 119 mimQTL miRNAs in the Oslo2 and TCGA cohorts. n) TargetScan [52] in silico predicted miRNA-target interactions for Global Methylation Alteration (GMA) score-correlated miRNAs and epigenetic regulator genes (conserved and non-conserved sites considered).