Skip to main content
. 2020 May 7;48(11):5814–5824. doi: 10.1093/nar/gkaa314

Table 2.

Performance (ARI and AMI) of different clustering methods from ten times analyses of the in-house human PBMC real dataset

ARI AMI
Mean (SD), [range] Mean (SD), [range]
Single source clustering - RNA only
K-means 0.547 (0.073), [0.429, 0.681] 0.613 (0.062), [0.474, 0.707]
TSCAN 0.476 (N/A), [N/A] 0.574 (N/A), [N/A]
SC3 0.561 (0.017), [0.540, 0.589] 0.606 (0.022), [0.560, 0.644]
DIMM-SC 0.563 (0.101), [0.366, 0.682] 0.598 (0.069), [0.488, 0.707]
Seurat 0.675 (N/A), [N/A] 0.732 (N/A), [N/A]
Single source clustering - protein only (complete list)
K-means 0.930 (0.120), [0.702, 0.989] 0.931 (0.072), [0.794, 0.968]
TSCAN 0.929 (N/A), [N/A] 0.881 (N/A), [N/A]
SC3 0.921 (0.039), [0.816, 0.951] 0.891 (0.048), [0.760, 0.930]
DIMM-SC 0.988 (0.005), [0.975, 0.993] 0.971 (0.014), [0.935, 0.983]
Seurat 0.996 (N/A), [N/A] 0.976 (N/A), [N/A]
Multi-source joint clustering
K-means 0.718 (0.102), [0.562, 0.805] 0.755 (0.057), [0.659, 0.803]
TSCAN 0.563 (N/A), [N/A] 0.619 (N/A), [N/A]
SC3 0.798 (0.108), [0.620, 0.938] 0.788 (0.068), [0.690, 0.881]
MV-NMF 0.861 (0.115), [0.710, 0.989] 0.871 (0.086), [0.780, 0.969]
PC-MSC 0.703 (0.094), [0.573, 0.950] 0.798 (0.048), [0.705, 0.909]
BREM-SC 0.966 (0.025), [0.917, 0.985] 0.944 (0.039), [0.869, 0.970]