KSHV regulates expression of the miR-30 family, which is predicted to target DLL4. (A) Heatmap representing relative changes in expression of hsa-miR-30 family members in LECs after KSHV infection. Red and yellow denote low and high expression, respectively. Four replicates of LECs and KLECs are shown. Two to 3 probes are shown for each member of the miR-30 family. Probes for hsa-miR-30b and hsa-miR-30c showed significant changes in expression (**Q < .01; ***Q < .001). Original GEM data from Lagos et al.²⁰  (B) Down-regulation of miR-30b and miR-30c in KLECs, confirmed by qRT-PCR (means + SEM, n = 3). Expression is relative to LECs. Differences between LECs and KLECs were significant (**P < .01). (C) Complementarity between miR-30 family members and the DLL4 3′UTR. Black lines indicate canonical Watson and Crick base-pairing, gray lines indicate G:U wobbles. The predicted target site within the DLL4 3′UTR, positions 59 to 66, is shown in red; miR-30 seed region is shown in green. (D) Heatmap table displaying the correlation coefficient R between expression of DLL4 and each member of the miR-30 family in the indicated tumor types. GEM and RNAseq data were obtained from The Cancer Genome Atlas (www.cancergenome.nih.gov) data portal (www.tcga-data.nci.nih.gov), as described in “Methods.” Because of the large difference in the number of replicates for each tumor type, the significance of R was calculated for each mir-30 versus DLL4 combination. The values and fill-in color indicate the degree of negative correlation. Nonsignificant correlations are grayed out; significant correlations are marked with an asterisk (*P < .05; **P < .01). Tumor types included are breast invasive carcinoma (BRCA), colon adenocarcinoma (COAD), glioblastoma multiforme (GBM), renal clear cell carcinoma (RCC), lung adenocarcinoma (LUAD), rectum adenocarcinoma (READ), and ovarian serous cystadenocarcinoma (OV).