Additional file 3 of An integrative gene expression signature analysis identifies CMS4 KRAS-mutated colorectal cancers sensitive to combined MEK and SRC targeted therapy

Additional file3. Fig S1-S11.pdf. Fig. S1.The 5-gene dasatinib sensitivity (Dasa-S) signature score predicted the trendof dasatinib sensitivity in multiple CRC cell lines (n=50). Fig.S2. The EMT genes were strongly associated with regionalmetastasis and disease recurrence in 2202 CRC tumors. Fig. S3.The CMS4 subtype was strongly correlated with the 13-gene MEKi“bypass”-resistance (13-gene BP), PC1, EMT, SRC activation and 5-gene Dasa-Ssignature scores in 1485 primary CRC tumors. Fig.S4. The CMS4 subtype was strongly correlated with the 13-geneMEKi “bypass”-resistance (13-gene BP), PC1, EMT, SRC activation and 5-geneDasa-S signature scores in 764 metastatic CRC tumors. Fig. S5.Scatter plots of Hu-Lgr5-ISC, Hu-EphB2-ISC, Hu-Late TA, and Hu-Proliferation vsEMT signature scores, respectively in Marisa 585 CRCs. Fig. S6.Scatter plots of Hu-Lgr5-ISC, Hu-EphB2-ISC, Hu-Late TA, and Hu-Proliferation vsEMT signature scores in TCGA 677 CRCs. Fig. S7.Spearman correlation heatmap of CMS1-4* scores, signature scores andEMT-associated genes in Marisa 585 CRC tumors. Fig.S8. Spearman correlation heatmap of CMS1-4* scores, signaturescores and EMT-associated genes in TCGA 677 CRC tumors. Fig. S9. Nodistinct association of MUT vs WT APC/TP53 tumors with the CMS subtypes. Fig. S10. Correlationanalysis of 154 CRC cell lines and in vitro drug treatment of HCT116 cells withMEKi + SRCi in CSC vs. non-CSC media. Fig.S11. In vitro drug treatment of LIM2405and HT29 cells with MEKi + SRCi in CSC vs. non-CSC media.