Fig. 5From: Evaluation of cell-cell interaction methods by integrating single-cell RNA sequencing data with spatial informationEvaluation results in the human SCC dataset. a Spatial cell type annotation result of ST data of sample P2_rep2 in the human SCC dataset. Color indicates the spot’s cell type. b The box plot of tools’ average distance enrichment scores (DES) evaluated using all samples (P2_rep2, P5_rep3, P10_rep1) in the human SCC dataset, sorted by the median DES. c The box plots of tools’ average relative accuracy metrics (F1 score, precision, recall) evaluated using all samples (P2_rep2, P5_rep3, P10_rep1) in the human SCC dataset, all sorted by the median F1 score. d The heatmap shows the average common interaction number between each tool’s result per cell type in the sample P2_rep2. Color displays the interaction number. The bar plot on the top shows the average interaction number predicted by each tool per cell type pair. The actual number is labeled on the top of each bar. e The dot plot of top 10 most common interactions among all tools in the sample P2_rep2, sorted by interaction’s common count. The dot size indicates the top rank percentage of the interaction in corresponding tool’s result. The dot color reflects interaction’s spatial distance tendency in the sample P2_rep2. The CCI tools are grouped by their model typesBack to article page