Fig. 4
From: Benchmarking principal component analysis for large-scale single-cell RNA-sequencing
Clustering accuracy comparison. Clustering accuracy was evaluated by the adjusted Rand index (ARI) of the result of Louvain clustering. Multiple principal component analysis (PCA) implementations were performed for PBMCs (102 cells), Pancreas (103 cells), BrainSpinalCord (105 cells), and Brain datasets (106 cells); Louvain clustering was performed for the PCA results. For each PCA result, Louvain clustering calculations were performed ten times and the average values were calculated. The cluster labels are the same as those of the respective original papers