Fig. 3
From: A global database for modeling tumor-immune cell communication
The predicted interactions stored in TICCom. (a) The number of samples in each cancer type from different bulk RNA-seq resources. The datasets with serial numbers containing MTAB or GSE were obtained from the EMBL-EBI Expression Atlas. (b) The number of predicted verified tumor-immune cell direct interactions and that of predicted ligand-receptor interactions in bulk RNA-seq data from 25 cancer types. The significant p values of TCGA and ICGC cancer were set at 0.05; however, the significant p value of EMBL was set at 0.1 because of the smaller number of samples. (c) The bubble plot showed the number of predicted ligand-receptor interactions that occurred in each cancer and each immune cell inferred by five algorithms based on 32 scRNA-seq datasets from 13 cancer types and the integrated ligand-receptor interactions. The bar plot showed the number of cells in each cancer. The size of the bubble indicated the number of predicted ligand-receptor interactions.
