Skip to main content

Table 1 Performance on eight PKFI sets

From: Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties

Group

Family

Members

Data size

ACC

MCC

AUROC

Positive

Negative

TK

EGFR

4

620

808

0.85

0.70

0.92

JAK

4

1347

877

0.84

0.64

0.91

CAMK

PIM

3

688

628

0.84

0.68

0.91

PKD

3

67

462

0.92

0.16

0.91

AGC

AKT

3

394

772

0.90

0.79

0.94

PKG

2

51

396

0.89

0.31

0.91

CMGC

GSK

2

243

657

0.75

0.28

0.81

CLK

4

178

418

0.72

0.20

0.79