Fig. 2

Single-cell RNA Sequencing (scRNA-Seq) data reveals the vital functions of FOXF1 and MFAP4 in cell adhesion and the metastasis of LUAD cells. (a) Using tSNE algorithm to divide all cells into 20 clusters. (b) The tSNE plot shows the sources of all cells. (c) The bubble plot shows the expression of marker genes used for cell annotation. (d) All cells are preliminarily divided into immune cells, epithelial cells and stromal cells. (e) Copy number variation (CNV) in epithelial cells and the stromal cells in normal lung tissues are used as reference cells. (f) Kmeans algorithm clustering all epithelial cells into 7 clusters according to CNV level. (g,h) Using tSNE algorithm to reduce the dimensionality of epithelial cells with high CNV level (LUAD tumor cells, Kmeans 3,4,5,6,7) into 7 cell clusters. (i) The expression level of FOXF1 and MFAP4 in LUAD tumor cells from different sources. (j) Functional enrichment analysis divides LUAD tumor cells into different functional modules. (k,l) Pseudotime analysis of 7 cell clusters calculated by tSNE algorithm. The tSNE plot was generated from “Seurat” package (version 4.4.0, https://github.com/satijalab/seurat) in R. The “ggplot2” (version 3.5.0, https://github.com/tidyverse/ggplot2) package in R was used to plot the heatmap and the bubble plot.