Fig. 4
From: A global database for modeling tumor-immune cell communication
The improvement of TICCom for prediction accuracy. (a) The number of experimentally-verified TIC (ev-TIC) communications in seven human resources and the integrated LR interactions. (b-e) The results of gene set enrichment analysis. To further illustrate the improvement of TICCom in prediction accuracy, we applied different algorithms, including ICELLNET, iTALK-top, NicheNet, and CellTalker, to infer cell-cell communication based on the integrated ligand-receptor interactions and iTALKDB LR interaction dataset in basal cell carcinoma (GSE12381365). The prediction result integrated from predicted cell-cell interactions inferred by these algorithms was used as the gene set. The prediction result of a single method was used as the ordered gene list, ranked by the communication score. (f) The percentage of top 50% of the cell-cell communication inferred from a single method occupied in the integrated result.
