Table 1 Comparison of signalling entropy to SLICE and StemID as measures of differentiation potency in scRNA-Seq and bulk RNA-Seq data sets.

From: Single-cell entropy for accurate estimation of differentiation potency from a cell’s transcriptome

Data set

 

Signalling entropy

SLICE

StemID

scRNA-Seq

 Chu1 (Pl>NonPl)

P

3e−132

∼1

3e−58

 

AUC

0.96

<0.5

0.79

 Chu2 (0 h>96 h)

P

2e−38

0.94

1e−22

 

AUC

0.97

<0.5

0.86

 Trapnell (0 h>72 h)

P

6e−9

0.0003

2e−10

 

AUC

0.74

0.65

0.75

 Treutlein (E14>Adult)

P

5e−27

6e−26

5e−27

 

AUC

1

0.998

1

Bulk RNA-Seq

 Chu3 (Pl>NonPl)

P

4e−5

0.001

0.76

 

AUC

0.99

0.90

<0.5

  1. Table lists one-tailed Wilcoxon rank-sum test P values and associated (one-tailed) AUCs, testing whether entropy is higher in the pluripotent or multipotent cells compared to the less potent cells in various scRNA-Seq and bulk RNA-Seq data sets. In Chu1, the comparison is between pluripotent (hESCs, n=374, Pl) and non-pluripotent (n=644, NonPl) single cells. In Chu2, the comparison is between hESCs (0 h, n=92) and definite endoderm progenitors sampled 96 h later (n=188). In Trapnell, the comparison is between human myoblasts (0 h, n=96) and differentiated skeletal muscle cells (72 h, n=84). In Treutlein, the comparison is between early lung progenitors (E14, n=45) and mature alveolar cells (n=46). In Chu3, the comparison is between bulk hESCs (n=7) and non-pluripotent samples (n=12). In bold-face we indicate the highest or relatively highest AUC values.
Back to article page