12920_2020_682_MOESM4_ESM.pptx (2.66 MB)
Download file

Additional file 4 of A pan-cancer study of class-3 semaphorins as therapeutic targets in cancer

Download (2.66 MB)
posted on 03.04.2020, 03:49 by Xiaoli Zhang, Brett Klamer, Jin Li, Soledad Fernandez, Lang Li
Additional file 4: Figure S1. (A) Histogram to show the number of samples for primary tumor and adjacent normal tissues in each cancer type and the sample number for blood cancer AML. (B) Histogram to show number of death event, number of censored, and number of not available for the overall survival analysis for each cancer type. Figure S2. Expression levels of SEMA3 genes in cancerous and adjacent normal tissues for all 31 cancer types. Boxplots represent the distribution of the SEMA3 gene expression levels (log2[RSEM normalized values relative to TBP]) in primary tumour and normal tissues (if available) of different cancer types for each of the SEMA3 genes. The band inside the box is the median expression values for the gene. Comparisons between normal and tumour expression values were performed with linear mixed effects models. A p-value< 0.009 after controlling 1 false positive among all the tests was considered as significance (53). Figure S3. Kaplan-Meier survival curves to show the correlation between SEMA3 gene expression and overall survival of patients with kidney clear cell carcinoma (KIRC). Gene expression was dichotomized into “Low” and “High” based on the median expression of each specific gene in KIRC and overall survival was tested between patients with low and high gene expression using log-rank tests. Worthnoting is that in Figure S3. the p-values are different from that of the Cox-proportional hazard model where gene expression was used as a continuous variable and all 7 genes were significantly associated with overall survival of KIRC. Figure S4. (A). Kaplan-Meier survival curve to show the overall survival difference among the six immune subtypes across all cancer types. (B) and (C). Correlation matrix plots to show the association between SEMA3 gene expression and immune scores (B), and Estimate scores (C), of 22 different cancer types based on ESTIMATE algorithm. Spearman correlation was used for testing. The size of the dots stands for the absolute value of the correlation coeffcients. The bigger the size is, the higher the correlation is (higher absolute correlation coefficient). Figure S5. (A). Boxplots to show the distribution of SEMA3 gene expression across NCI-60 cell lines. (B). Heatmap to show SEMA3 gene expression within each individual cell lines using NCI-60 cell line data. (C). Boxplots to show the sensitivity score distribution of the FDA approved chemotherapy drugs that are significantly associated with SEMA3 gene expression. Figure S6. (A). Boxplot to show the distribution of SEMA3 gene expression comparing primary tumor to adjacent normal in breast cancer. (B). Correlation plot to show the correlation of gene expression among the 7 SEMA3 family members in breast cancer. Figure S7. Boxplots to show the association of SEMA3 gene expressions with molecular subtypes of BLCA, HNSC, KIRC, and OV tested with ANOVA (P < .0001).