Fig. 3: Common pathways are affected by TTFields across tumor types.

A Enrichment map for 48-h TTFields-treatment datasets, representing commonly affected Reactome pathways. Analysis was done according to ActivePathways package, and the pathways clustered according to Reactome head pathways (header size indicative of the size of the cluster). Each node represents a pathway that was significantly altered in at least 5 datasets, with the pie colors depicting the different contributing datasets (see legend) and the pie size indicative of the common significance (Brown procedure). The edges correspond to the number of genes common to the pathways. The clustered circles were generated through AutoAnnotate. Only the top 5 pathways are shown. The complete map may be found in Fig. S2. B GSEA heatmap for 48-h TTFields-treatment datasets performed with MsigDB hallmark gene sets, showing pathways which were significant for at least 5 datasets. Normalized Enrichment Score (NES): Red = upregulated, Blue = downregulated. False Discovery Rate (FDR): * FDR ≤ 0.05, ** FDR ≤ 0.01, *** FDR ≤ 0.001. Cell lines from each tumor type are labeled with a different color. C Dot plot of IPA canonical pathways changed in response to TTFields application. A threshold p-value ≤ 0.05 and absolute log fold change ≥0.415 was applied. Only pathways that had a significance of BH p-value ≤ 0.05 in at least 5 datasets are shown. The circle color represents the Z-score. Red = upregulated, Blue = downregulated. The circle size corresponds to the -log BH p-value. The order of the datasets is according to the hierarchical clustering in (B).