Figure 1. Characterization of single-cell RNA sequencing from 121 cells and screening of marker genes. (A, B) Quality control of scRNA-seq for three cell sub-populations. We filtered out the cells with poor quality and analyzed the positive associations between detected gene counts and sequencing depth. (C) we identified the gene symbols with significant difference across cells and drawn the characteristic variance diagram. (D, E) The principal component analysis (PCA), a linear dimensionality reduction method, was ultilized to identify the significantly available dimensions of data sets with estimated P value. Accordingly, we classified the cell groups into three categories. (F) Based on available significant components from PCA, we conducted another nonlinear dimensionality reduction, TSNE algorithm, to successfully divided the cells into two clusters, in accordance with actual cell types. (G) Differential analysis with logFC =0.5 and adjPval =0.05 was constructed between two clusters to identify significant marker genes and we exhibited the top 20 in heatmap package. (H) Cell annotations and trajectory analysis revealed the tendency curve from primary RCC to metastatic ones, indicating the genomic alternations between them.