Fig. 6

Bioinformatics analysis of the correlation between CCL8 and FMD. A Differential gene expression (GEO database, dataset GSE50712, Control vs. EVE-treated breast tumor bearing mice) is determined using GEO2R; the log-fold-change is used to rank the genes. B–D The expression of CCL8, TGFB1, and HGF in invasive breast cancer samples from the TCGA database. E KEGG enrichment analysis for data from the GEO database (dataset GSE37956, control vs. bevacizumab treated glioblastoma xenograft tumor). F The expression of CCL8, TGFB1, PKG1, LDHA, and HIF1A in GEO database (dataset GSE37956, control vs. bevacizumab treated glioblastoma xenograft tumor). G, H Correlation between CCL8, TGFBI and immune cell infiltration in invasive breast cancer samples from the TCGA database. I GSVA analysis of correlation between CCL8, hypoxia (Gene set from GESA M34030), glycolysis (Gene set from GESA M5113), glucose starvation (Gene set from GESA M15497), mTOR (Gene set from GESA M2672), macrophage (Gene set from GESA M39708), M2 macrophage (Gene set from GESA M8527), M1 macrophage (Gene set from GESA M6586) in invasive breast cancer samples from the TCGA database