Table 1 Performance of clustering across ten times analyses for three real datasets

From: A Bayesian mixture model for clustering droplet-based single-cell transcriptomic data from population studies

Method

Mean_P

SD_P

Range_P

Mean_L

SD_L

Range_L

Mean_S

SD_S

Range_S

MNN+K-means

0.379

0.083

(0.283–0.485)

0.662

0.066

(0.596–0.815)

0.597

0.075

(0.461–0.676)

MNN+TSCAN

0.373

NA

NA

0.720

NA

NA

0.553

NA

NA

MNN+SC3

0.348

0.084

(0.266–0.511)

0.640

0.061

(0.556–0.687)

0.517

0.034

(0.436–0.557)

MNN+Seurat

0.325

NA

NA

0.749

NA

NA

0.647

NA

NA

CCA+K-means

0.414

0.056

(0.307–0.464)

0.695

0.114

(0.505–0.883)

0.619

0.129

(0.424–0.737)

CCA+TSCAN

0.210

NA

NA

0.611

NA

NA

0.398

NA

NA

CCA+SC3

0.145

0.052

(0.051–0.215)

0.610

0.068

(0.531–0.708)

0.369

0.071

(0.277–0.488)

CCA+Seurat

0.468

NA

NA

0.729

NA

NA

0.702

NA

NA

DIMM-SC

0.333

0.071

(0.302–0.529)

0.809

0.030

(0.742–0.868)

0.715

0.045

(0.671–0.779)

BAMM-SC

0.487

0.056

(0.362–0.532)

0.882

0.042

(0.764–0.910)

0.762

0.032

(0.717–0.843)

  1. Columns Mean_P, SD_P, and Range_P were calculated from human PBMC dataset. Columns Mean_L, SD_L, and Range_L were calculated from mouse lung dataset. Columns Mean_S, SD_S, and Range_S were calculated from human skin dataset.